| Original Filename: | WIL027_Williams_S_SA100_24-25_1778504076.pdf |
| Upload Date: | May 11, 2026 at 01:00 PM |
| Total Pages: | 31 |
| Processing Status: | Partial |
| Total Documents: | 15 |
| Corporation Taxs: | 7 |
| Financial Statements: | 5 |
| Property Statements: | 2 |
| P32 Payrolls: | 1 |
Corporation Tax
95%Corporation Tax
90%Corporation Tax
95%Corporation Tax
95%Corporation Tax
95%Corporation Tax
90%Corporation Tax
95%Corporation Tax
95%Corporation Tax
90%Corporation Tax
90%Corporation Tax
95%Financial Statement
95%Corporation Tax
95%Corporation Tax
95%Financial Statement
95%Corporation Tax
95%Property Statement
95%Property Statement
90%Corporation Tax
95%Corporation Tax
90%Corporation Tax
95%Financial Statement
85%Unknown
80%P32 Payroll
95%Corporation Tax
85%Corporation Tax
90%Unknown
80%Financial Statement
95%Financial Statement
95%Property Statement
95%Property Statement
95%| Document Type | Pages | Confidence | Processing Status | Created | Actions |
|---|---|---|---|---|---|
| Corporation Tax | 1-2 (2 pages) | 95% | Processed | 2026-05-11 13:00 | No view available |
| Corporation Tax | 3-11 (9 pages) | 95% | Processed | 2026-05-11 13:00 | No view available |
| Financial Statement | 12 (1 page) | 95% | Processed | 2026-05-11 13:00 | No view available |
| Corporation Tax | 13-14 (2 pages) | 95% | Processed | 2026-05-11 13:00 | No view available |
| Financial Statement | 15 (1 page) | 95% | Processed | 2026-05-11 13:00 | No view available |
| Corporation Tax | 16 (1 page) | 95% | Processed | 2026-05-11 13:00 | No view available |
| Property Statement | 17-18 (2 pages) | 92% | Failed | 2026-05-11 13:00 | Processing... |
| Corporation Tax | 19-20 (2 pages) | 95% | Processed | 2026-05-11 13:00 | No view available |
| Corporation Tax | 21 (1 page) | 95% | Processed | 2026-05-11 13:00 | No view available |
| Financial Statement | 22 (1 page) | 85% | Processed | 2026-05-11 13:00 | No view available |
| P32 Payroll | 24 (1 page) | 95% | Processed | 2026-05-11 13:00 | No view available |
| Corporation Tax | 25-26 (2 pages) | 88% | Processed | 2026-05-11 13:00 | No view available |
| Financial Statement | 28 (1 page) | 95% | Processed | 2026-05-11 13:00 | No view available |
| Financial Statement | 29 (1 page) | 95% | Processed | 2026-05-11 13:00 | No view available |
| Property Statement | 30-31 (2 pages) | 95% | Failed | 2026-05-11 13:00 | Processing... |
{
"document_groups": [
{
"avg_confidence": 0.9318181818181818,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "corporation_tax",
"end_page": 11,
"pages": [
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a Tax Return for a specific year, and mentions of Self Assessment Online Service, IRmark number, and tax repayment claims. These are key indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWMmDnPajuE7grpXycO4On8OYk0\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a Tax Return for a specific year, and mentions of Self Assessment Online Service, IRmark number, and tax repayment claims. These are key indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504374, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=209, prompt_tokens=1383, total_tokens=1592, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(audio_tokens=0, cached_to"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HM Revenue \u0026 Customs, a Tax Return for a specific year, and mentions of Self Assessment Online Service, IRmark number, and tax repayment claims. These are key indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a Tax Return for a specific year, and mentions of Self Assessment Online Service, IRmark number, and tax repayment claims. These are key indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.13270588235294117,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.1,
"total_score": 0.13270588235294117
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "corporation_tax"
}
},
"page_number": 1,
"text_length": 1899
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return, schedule reports, rate band analysis, client payment advice, payments on account, and other tax-related terms. These are strong indicators of a corporation tax document, likely related to a tax return or computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWNuOli5vXtiy4ynfGwNrhIHYkU\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to a tax return, schedule reports, rate band analysis, client payment advice, payments on account, and other tax-related terms. These are strong indicators of a corporation tax document, likely related to a tax return or computation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504375, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=212, prompt_tokens=1047, total_tokens=1259, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetai"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to a tax return, schedule reports, rate band analysis, client payment advice, payments on account, and other tax-related terms. These are strong indicators of a corporation tax document, likely related to a tax return or computation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return, schedule reports, rate band analysis, client payment advice, payments on account, and other tax-related terms. These are strong indicators of a corporation tax document, likely related to a tax return or computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.038461538461538464,
"pattern_score": 0.0,
"total_score": 0.023076923076923078
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 2,
"text_length": 249
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax return, tax year, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. Additionally, it includes details about deadlines for filing tax returns and penalties for late submissions, which are specific to tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWPM0Ixg35qMRqOLaSLmh1SzSNZ\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax return, tax year, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. Additionally, it includes details about deadlines for filing tax returns and penalties for late submissions, which are specific to tax-related documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504377, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=224, prompt_tokens=1727, total_tokens=1951, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax return, tax year, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. Additionally, it includes details about deadlines for filing tax returns and penalties for late submissions, which are specific to tax-related documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax return, tax year, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. Additionally, it includes details about deadlines for filing tax returns and penalties for late submissions, which are specific to tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.1807058823529412,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.2,
"total_score": 0.1807058823529412
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 3,
"text_length": 2696
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to a tax return, tax reference number, taxable income, capital gains tax, self-employment income, UK property income, and other tax-related information. These are strong indicators of a corporation tax document or a tax return form.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWQS2sBU9FfcUD8XNWTMQ8LIFEw\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to a tax return, tax reference number, taxable income, capital gains tax, self-employment income, UK property income, and other tax-related information. These are strong indicators of a corporation tax document or a tax return form.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504378, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1730, total_tokens=1944, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDeta"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to a tax return, tax reference number, taxable income, capital gains tax, self-employment income, UK property income, and other tax-related information. These are strong indicators of a corporation tax document or a tax return form.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to a tax return, tax reference number, taxable income, capital gains tax, self-employment income, UK property income, and other tax-related information. These are strong indicators of a corporation tax document or a tax return form.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 4,
"text_length": 4065
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027SA 100\u0027. These are key indicators of a tax-related document, specifically a corporation tax return or related form. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return document, which falls under the corporation_tax category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWSwbOpBxxycfZazovwqIGslk6D\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Taxed UK interest\\\u0027, \\\u0027Untaxed UK interest\\\u0027, \\\u0027Foreign dividends\\\u0027, \\\u0027Taxable Incapacity Benefit\\\u0027, and \\\u0027SA 100\\\u0027. These are key indicators of a tax-related document, specifically a corporation tax return or related form. The presence of \\\u0027SA 100\\\u0027 further confirms that this is a self-assessment tax return document, which falls under the corporation_tax category.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504380, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=256, prompt_tokens=1635, total_tokens="
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027SA 100\u0027. These are key indicators of a tax-related document, specifically a corporation tax return or related form. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return document, which falls under the corporation_tax category.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027SA 100\u0027. These are key indicators of a tax-related document, specifically a corporation tax return or related form. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return document, which falls under the corporation_tax category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.15,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.125,
"total_score": 0.15
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.030303030303030304,
"total_score": 0.03681818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 5,
"text_length": 2229
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and other tax-related allowances, which are typically part of a corporation tax return or related documentation. The presence of a tax reference number and the mention of \u0027SA 100\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWT3zve4Bm6jtLYusCpQHzDNIJc\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and other tax-related allowances, which are typically part of a corporation tax return or related documentation. The presence of a tax reference number and the mention of \\\u0027SA 100\\\u0027 further supports this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504381, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=219, prompt_tokens=1598, total_tokens=1817, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_token"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax reliefs, pension contributions, charitable giving, and other tax-related allowances, which are typically part of a corporation tax return or related documentation. The presence of a tax reference number and the mention of \u0027SA 100\u0027 further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and other tax-related allowances, which are typically part of a corporation tax return or related documentation. The presence of a tax reference number and the mention of \u0027SA 100\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.030303030303030304,
"total_score": 0.03681818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 6,
"text_length": 2174
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWU39VxwWRCh2RPtZQkxPtgQ2Eh\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Student Loan and Postgraduate Loan repayments\\\u0027, \\\u0027High Income Child Benefit Charge\\\u0027, and \\\u0027Marriage Allowance\\\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \\\u0027SA 100\\\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504382, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usag"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11454545454545456,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.09090909090909091,
"total_score": 0.11454545454545456
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.030303030303030304,
"total_score": 0.06681818181818183
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "loan_statement"
}
},
"page_number": 7,
"text_length": 2181
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax refunds, and specific tax years (e.g., 2024-25, 2025-26). It also mentions terms like \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary,\u0027 which are indicative of a corporation tax document. Additionally, the document includes a client reference number and tax reference, which are typical identifiers in tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWW042fG8HN0NpTROfUb7Q24NNf\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax refunds, and specific tax years (e.g., 2024-25, 2025-26). It also mentions terms like \\\u0027Self Assessment Statement,\\\u0027 \\\u0027Tax Reference,\\\u0027 and \\\u0027Tax calculation summary,\\\u0027 which are indicative of a corporation tax document. Additionally, the document includes a client reference number and tax reference, which are typical identifiers in tax-related documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504384, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=248, prompt_tokens=1673, total_tokens=1921, completion_tokens_detai"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax returns, tax calculations, tax refunds, and specific tax years (e.g., 2024-25, 2025-26). It also mentions terms like \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary,\u0027 which are indicative of a corporation tax document. Additionally, the document includes a client reference number and tax reference, which are typical identifiers in tax-related documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax refunds, and specific tax years (e.g., 2024-25, 2025-26). It also mentions terms like \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary,\u0027 which are indicative of a corporation tax document. Additionally, the document includes a client reference number and tax reference, which are typical identifiers in tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.1392857142857143,
"features": {
"bank_statement": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.0,
"total_score": 0.10588235294117647
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.14285714285714285,
"total_score": 0.1392857142857143
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.030303030303030304,
"total_score": 0.05181818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 8,
"text_length": 2552
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax adviser, a tax reference number, and mentions \u0027Income from property\u0027 and \u0027SA 100\u0027, which are indicative of tax-related documentation. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Client Ref\u0027, and \u0027SA 100\u0027 strongly suggests this is related to corporation tax or self-assessment tax returns.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWYjJe69dwb6PhFCWwpZJ2xCmr8\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to a tax adviser, a tax reference number, and mentions \\\u0027Income from property\\\u0027 and \\\u0027SA 100\\\u0027, which are indicative of tax-related documentation. Additionally, the presence of terms like \\\u0027Tax Reference\\\u0027, \\\u0027Client Ref\\\u0027, and \\\u0027SA 100\\\u0027 strongly suggests this is related to corporation tax or self-assessment tax returns.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504386, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=234, prompt_tokens=1187, total_tokens=1421, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, r"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to a tax adviser, a tax reference number, and mentions \u0027Income from property\u0027 and \u0027SA 100\u0027, which are indicative of tax-related documentation. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Client Ref\u0027, and \u0027SA 100\u0027 strongly suggests this is related to corporation tax or self-assessment tax returns.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax adviser, a tax reference number, and mentions \u0027Income from property\u0027 and \u0027SA 100\u0027, which are indicative of tax-related documentation. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Client Ref\u0027, and \u0027SA 100\u0027 strongly suggests this is related to corporation tax or self-assessment tax returns.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 9,
"text_length": 635
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return (SA100), tax reference number, and a declaration section, which are key indicators of a corporation tax document. Additionally, it includes instructions for filling out and signing the tax return, which aligns with the characteristics of corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWaQ53wUyuAXipTGFFEeD9AevV8\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to a tax return (SA100), tax reference number, and a declaration section, which are key indicators of a corporation tax document. Additionally, it includes instructions for filling out and signing the tax return, which aligns with the characteristics of corporation tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504388, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=218, prompt_tokens=1262, total_tokens=1480, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tok"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to a tax return (SA100), tax reference number, and a declaration section, which are key indicators of a corporation tax document. Additionally, it includes instructions for filling out and signing the tax return, which aligns with the characteristics of corporation tax documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return (SA100), tax reference number, and a declaration section, which are key indicators of a corporation tax document. Additionally, it includes instructions for filling out and signing the tax return, which aligns with the characteristics of corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.0,
"total_score": 0.045
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 10,
"text_length": 1070
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, Unique Taxpayer Reference (UTR), business details, turnover, and other income details, which are key indicators of a corporation tax document. Additionally, the form is labeled as \u0027SA103F,\u0027 which is a self-employment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWbUjNSEKMQ7P7GxrzNz9Y26nzb\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax year, Unique Taxpayer Reference (UTR), business details, turnover, and other income details, which are key indicators of a corporation tax document. Additionally, the form is labeled as \\\u0027SA103F,\\\u0027 which is a self-employment tax return form used in the UK.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504389, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=229, prompt_tokens=1570, total_tokens=1799, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), p"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax year, Unique Taxpayer Reference (UTR), business details, turnover, and other income details, which are key indicators of a corporation tax document. Additionally, the form is labeled as \u0027SA103F,\u0027 which is a self-employment tax return form used in the UK.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, Unique Taxpayer Reference (UTR), business details, turnover, and other income details, which are key indicators of a corporation tax document. Additionally, the form is labeled as \u0027SA103F,\u0027 which is a self-employment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.21741176470588236,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.23529411764705882,
"pattern_score": 0.1,
"total_score": 0.21741176470588236
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.15789473684210525,
"pattern_score": 0.0,
"total_score": 0.11368421052631578
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.0,
"total_score": 0.13
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 11,
"text_length": 1871
}
],
"start_page": 1
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "financial_statement",
"end_page": 12,
"pages": [
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains detailed financial information related to business expenses, including categories such as wages, salaries, travel expenses, rent, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWcpe1HU5hkrHzH93RahS30vFmZ\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains detailed financial information related to business expenses, including categories such as wages, salaries, travel expenses, rent, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504390, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=229, prompt_tokens=1585, total_tokens=1814, completion_tokens_details=CompletionToken"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains detailed financial information related to business expenses, including categories such as wages, salaries, travel expenses, rent, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains detailed financial information related to business expenses, including categories such as wages, salaries, travel expenses, rent, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.03529411764705882
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.0,
"total_score": 0.04
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 12,
"text_length": 1488
}
],
"start_page": 12
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "corporation_tax",
"end_page": 14,
"pages": [
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related terms such as \u0027Tax Reference\u0027, \u0027Net profit or loss\u0027, \u0027Tax allowances for vehicles and equipment\u0027, \u0027capital allowances\u0027, and \u0027taxable profit or loss\u0027. Additionally, it includes specific tax-related calculations and references to HMRC forms (e.g., SA103F). These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWePuli9M2fjEwYtcYmIbpwwhn0\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related terms such as \\\u0027Tax Reference\\\u0027, \\\u0027Net profit or loss\\\u0027, \\\u0027Tax allowances for vehicles and equipment\\\u0027, \\\u0027capital allowances\\\u0027, and \\\u0027taxable profit or loss\\\u0027. Additionally, it includes specific tax-related calculations and references to HMRC forms (e.g., SA103F). These are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504392, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=236, prompt_tokens=1724, total_tokens=1960, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0,"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related terms such as \u0027Tax Reference\u0027, \u0027Net profit or loss\u0027, \u0027Tax allowances for vehicles and equipment\u0027, \u0027capital allowances\u0027, and \u0027taxable profit or loss\u0027. Additionally, it includes specific tax-related calculations and references to HMRC forms (e.g., SA103F). These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related terms such as \u0027Tax Reference\u0027, \u0027Net profit or loss\u0027, \u0027Tax allowances for vehicles and equipment\u0027, \u0027capital allowances\u0027, and \u0027taxable profit or loss\u0027. Additionally, it includes specific tax-related calculations and references to HMRC forms (e.g., SA103F). These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 13,
"text_length": 2548
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWgAUnNXIwHfliac8ATt1aC0Gka\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \\\u0027Self-employment (full) notes,\\\u0027 \\\u0027taxable profits,\\\u0027 and \\\u0027losses,\\\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504394, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=253, prompt_tokens=1729, total_tokens=1982, co"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.08470588235294117,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.08470588235294117
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 14,
"text_length": 2609
}
],
"start_page": 13
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "financial_statement",
"end_page": 15,
"pages": [
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital of a business. Specific terms such as \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 and \u0027Net business assets\u0027 are present, which are key indicators of a financial statement. There is no evidence of transactions, invoices, payroll, or other document types listed in the categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWhkkYAd0tc4PYchrKpvd8s5xUt\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital of a business. Specific terms such as \\\u0027Balance sheet,\\\u0027 \\\u0027Assets,\\\u0027 \\\u0027Liabilities,\\\u0027 and \\\u0027Net business assets\\\u0027 are present, which are key indicators of a financial statement. There is no evidence of transactions, invoices, payroll, or other document types listed in the categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504395, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=235, prompt_tokens=1539, total_tokens=1774, completion_tokens_details=CompletionTokensDetails(accepted_p"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a balance sheet with detailed information about assets, liabilities, and capital of a business. Specific terms such as \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 and \u0027Net business assets\u0027 are present, which are key indicators of a financial statement. There is no evidence of transactions, invoices, payroll, or other document types listed in the categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital of a business. Specific terms such as \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 and \u0027Net business assets\u0027 are present, which are key indicators of a financial statement. There is no evidence of transactions, invoices, payroll, or other document types listed in the categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.2374736842105263,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.2631578947368421,
"pattern_score": 0.1,
"total_score": 0.2374736842105263
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 15,
"text_length": 1430
}
],
"start_page": 15
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "corporation_tax",
"end_page": 16,
"pages": [
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and specific details about Class 2 and Class 4 NICs. It also includes a client reference and mentions self-employment profits, which are indicative of a tax-related document. The format and content align with HMRC tax documentation, specifically related to self-assessment forms.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWjA5AdDrEmvlMjbDNTbx5Ga7rA\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and specific details about Class 2 and Class 4 NICs. It also includes a client reference and mentions self-employment profits, which are indicative of a tax-related document. The format and content align with HMRC tax documentation, specifically related to self-assessment forms.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504397, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=240, prompt_tokens=1210, total_tokens=1450, completion_tokens_details=Completion"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and specific details about Class 2 and Class 4 NICs. It also includes a client reference and mentions self-employment profits, which are indicative of a tax-related document. The format and content align with HMRC tax documentation, specifically related to self-assessment forms.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and specific details about Class 2 and Class 4 NICs. It also includes a client reference and mentions self-employment profits, which are indicative of a tax-related document. The format and content align with HMRC tax documentation, specifically related to self-assessment forms.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 16,
"text_length": 764
}
],
"start_page": 16
},
{
"avg_confidence": 0.925,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"completion_date": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"purchase_price": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "property_statement",
"end_page": 18,
"pages": [
{
"confidence_score": 0.95,
"document_type": "property_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to UK property details, rental income, allowable expenses, and other property-related financial information. It also includes terms like \u0027property income,\u0027 \u0027rented out,\u0027 and \u0027furnished holiday lettings,\u0027 which are key indicators of a property statement. Additionally, the document is labeled as \u0027SA 105,\u0027 which is a supplementary page for property income in UK tax returns.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_price\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWkTzT1AIqCyt9UY9VmCM2dRtFZ\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"property_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to UK property details, rental income, allowable expenses, and other property-related financial information. It also includes terms like \\\u0027property income,\\\u0027 \\\u0027rented out,\\\u0027 and \\\u0027furnished holiday lettings,\\\u0027 which are key indicators of a property statement. Additionally, the document is labeled as \\\u0027SA 105,\\\u0027 which is a supplementary page for property income in UK tax returns.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null,\\n \"completion_date\": null,\\n \"property_address\": null,\\n \"purchase_price\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504398, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=Completio"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"completion_date": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"purchase_price": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to UK property details, rental income, allowable expenses, and other property-related financial information. It also includes terms like \u0027property income,\u0027 \u0027rented out,\u0027 and \u0027furnished holiday lettings,\u0027 which are key indicators of a property statement. Additionally, the document is labeled as \u0027SA 105,\u0027 which is a supplementary page for property income in UK tax returns.",
"raw_response": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to UK property details, rental income, allowable expenses, and other property-related financial information. It also includes terms like \u0027property income,\u0027 \u0027rented out,\u0027 and \u0027furnished holiday lettings,\u0027 which are key indicators of a property statement. Additionally, the document is labeled as \u0027SA 105,\u0027 which is a supplementary page for property income in UK tax returns.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_price\": null\n }\n}\n```"
},
"type": "property_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.21741176470588236,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.23529411764705882,
"pattern_score": 0.1,
"total_score": 0.21741176470588236
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.0,
"total_score": 0.06
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.0,
"total_score": 0.045
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 17,
"text_length": 2125
},
{
"confidence_score": 0.9,
"document_type": "property_statement",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to property income, property expenses, residential property finance costs, and property-related allowances. It also includes a client reference and tax reference, which are common in property-related financial documents. The presence of terms like \u0027Property income\u0027, \u0027Property expenses\u0027, and \u0027Residential property finance costs\u0027 strongly indicates that this is a property statement.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_sale_price\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWme6wSu95XTFKw81I7Zavyty4j\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"property_statement\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to property income, property expenses, residential property finance costs, and property-related allowances. It also includes a client reference and tax reference, which are common in property-related financial documents. The presence of terms like \\\u0027Property income\\\u0027, \\\u0027Property expenses\\\u0027, and \\\u0027Residential property finance costs\\\u0027 strongly indicates that this is a property statement.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null,\\n \"completion_date\": null,\\n \"property_address\": null,\\n \"purchase_sale_price\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504400, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"completion_date": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"purchase_sale_price": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to property income, property expenses, residential property finance costs, and property-related allowances. It also includes a client reference and tax reference, which are common in property-related financial documents. The presence of terms like \u0027Property income\u0027, \u0027Property expenses\u0027, and \u0027Residential property finance costs\u0027 strongly indicates that this is a property statement.",
"raw_response": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to property income, property expenses, residential property finance costs, and property-related allowances. It also includes a client reference and tax reference, which are common in property-related financial documents. The presence of terms like \u0027Property income\u0027, \u0027Property expenses\u0027, and \u0027Residential property finance costs\u0027 strongly indicates that this is a property statement.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_sale_price\": null\n }\n}\n```"
},
"type": "property_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 18,
"text_length": 2119
}
],
"start_page": 17
},
{
"avg_confidence": 0.9333333333333332,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "corporation_tax",
"end_page": 21,
"pages": [
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax calculation summary, Self Assessment, and specific tax-related terms such as \u0027Tax year\u0027, \u0027Unique Taxpayer Reference (UTR)\u0027, \u0027Class 2 NICs\u0027, \u0027Class 4 NICs\u0027, \u0027Capital Gains Tax\u0027, and \u0027Payments on account\u0027. These are clear indicators of a tax-related document, specifically a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWod4NzhIvZhmIaE5rXX0MFHiFl\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax calculation summary, Self Assessment, and specific tax-related terms such as \\\u0027Tax year\\\u0027, \\\u0027Unique Taxpayer Reference (UTR)\\\u0027, \\\u0027Class 2 NICs\\\u0027, \\\u0027Class 4 NICs\\\u0027, \\\u0027Capital Gains Tax\\\u0027, and \\\u0027Payments on account\\\u0027. These are clear indicators of a tax-related document, specifically a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504402, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=240, prompt_tokens=1689, total_tokens=1929, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax calculation summary, Self Assessment, and specific tax-related terms such as \u0027Tax year\u0027, \u0027Unique Taxpayer Reference (UTR)\u0027, \u0027Class 2 NICs\u0027, \u0027Class 4 NICs\u0027, \u0027Capital Gains Tax\u0027, and \u0027Payments on account\u0027. These are clear indicators of a tax-related document, specifically a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax calculation summary, Self Assessment, and specific tax-related terms such as \u0027Tax year\u0027, \u0027Unique Taxpayer Reference (UTR)\u0027, \u0027Class 2 NICs\u0027, \u0027Class 4 NICs\u0027, \u0027Capital Gains Tax\u0027, and \u0027Payments on account\u0027. These are clear indicators of a tax-related document, specifically a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.17505882352941174,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.1,
"total_score": 0.17505882352941174
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.09090909090909091,
"total_score": 0.11454545454545456
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.06060606060606061,
"total_score": 0.08863636363636364
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 19,
"text_length": 2286
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027tax due,\u0027 \u0027repayment,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWqZnFJjJjKYBDQCS06KiHPLKZA\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \\\u0027tax due,\\\u0027 \\\u0027repayment,\\\u0027 and \\\u0027self-assessment statement of account.\\\u0027 These are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504404, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=224, prompt_tokens=1326, total_tokens=1550, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptToken"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027tax due,\u0027 \u0027repayment,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027tax due,\u0027 \u0027repayment,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.07058823529411765,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "bank_statement"
}
},
"page_number": 20,
"text_length": 1359
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains tax-related information, including a computation summary, taxable income, tax rates, and National Insurance Contributions (NIC). It also references a tax year (2024/25) and includes a tax reference number, which are strong indicators of a tax-related document. The structure and content align closely with a tax return or computation document, which falls under the \u0027corporation_tax\u0027 category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWrZuhfzgpjLH8oZFjFHSmvS6Nx\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains tax-related information, including a computation summary, taxable income, tax rates, and National Insurance Contributions (NIC). It also references a tax year (2024/25) and includes a tax reference number, which are strong indicators of a tax-related document. The structure and content align closely with a tax return or computation document, which falls under the \\\u0027corporation_tax\\\u0027 category.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504405, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=242, prompt_tokens=1327, total_tokens=1569, completion_tokens_details=Completi"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains tax-related information, including a computation summary, taxable income, tax rates, and National Insurance Contributions (NIC). It also references a tax year (2024/25) and includes a tax reference number, which are strong indicators of a tax-related document. The structure and content align closely with a tax return or computation document, which falls under the \u0027corporation_tax\u0027 category.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains tax-related information, including a computation summary, taxable income, tax rates, and National Insurance Contributions (NIC). It also references a tax year (2024/25) and includes a tax reference number, which are strong indicators of a tax-related document. The structure and content align closely with a tax return or computation document, which falls under the \u0027corporation_tax\u0027 category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.038461538461538464,
"pattern_score": 0.0,
"total_score": 0.023076923076923078
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.030303030303030304,
"total_score": 0.08181818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 21,
"text_length": 1026
}
],
"start_page": 19
},
{
"avg_confidence": 0.85,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "financial_statement",
"end_page": 22,
"pages": [
{
"confidence_score": 0.85,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWtaSM0Enrox0EyJ4mju2XXqS1Q\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504407, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1970, total_tokens=2184, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prom"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 22,
"text_length": 1715
}
],
"start_page": 22
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "p32_payroll",
"end_page": 24,
"pages": [
{
"confidence_score": 0.95,
"document_type": "p32_payroll",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and \u0027Amounts Due To Be Paid Via Instalments,\u0027 which are indicative of HMRC P32 Employer Payment Records. Additionally, it includes specific payment dates and amounts, which align with tax obligations and employer payment records.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWv86pprGbOWXmHXaFaLHZhmEuU\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"p32_payroll\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to \\\u0027Client Payment Advice,\\\u0027 \\\u0027Payments Required,\\\u0027 \\\u0027Payment Dates,\\\u0027 and \\\u0027Amounts Due To Be Paid Via Instalments,\\\u0027 which are indicative of HMRC P32 Employer Payment Records. Additionally, it includes specific payment dates and amounts, which align with tax obligations and employer payment records.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504409, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=225, prompt_tokens=1135, total_tokens=1360, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rej"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and \u0027Amounts Due To Be Paid Via Instalments,\u0027 which are indicative of HMRC P32 Employer Payment Records. Additionally, it includes specific payment dates and amounts, which align with tax obligations and employer payment records.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and \u0027Amounts Due To Be Paid Via Instalments,\u0027 which are indicative of HMRC P32 Employer Payment Records. Additionally, it includes specific payment dates and amounts, which align with tax obligations and employer payment records.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "p32_payroll"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 24,
"text_length": 397
}
],
"start_page": 24
},
{
"avg_confidence": 0.875,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "corporation_tax",
"end_page": 26,
"pages": [
{
"confidence_score": 0.85,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of a tax computation or return. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting. These indicators suggest the document is related to corporation tax or a similar tax computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWxzYW1h4ckx0zuI5yoX3yj6eUx\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document references tax-related terms such as \\\u0027Tax Due After Notional Taxes Paid,\\\u0027 \\\u0027Class 4 NIC Due,\\\u0027 and \\\u0027Payments on Account,\\\u0027 which are indicative of a tax computation or return. The mention of \\\u0027Year ended 5 April 2025\\\u0027 aligns with tax year reporting. These indicators suggest the document is related to corporation tax or a similar tax computation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504411, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=238, prompt_tokens=1109, total_tokens=1347, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_t"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of a tax computation or return. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting. These indicators suggest the document is related to corporation tax or a similar tax computation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of a tax computation or return. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting. These indicators suggest the document is related to corporation tax or a similar tax computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 25,
"text_length": 374
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document references \u0027Class 4 National Insurance Charge,\u0027 \u0027Self-Employment Profit,\u0027 and specific tax-related calculations, which are indicative of a tax-related document. The structure and terminology align with corporation tax or self-assessment tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWz5JTXzgHRKHdmIxSq4HzFwhzG\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document references \\\u0027Class 4 National Insurance Charge,\\\u0027 \\\u0027Self-Employment Profit,\\\u0027 and specific tax-related calculations, which are indicative of a tax-related document. The structure and terminology align with corporation tax or self-assessment tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504413, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=210, prompt_tokens=1121, total_tokens=1331, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDet"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document references \u0027Class 4 National Insurance Charge,\u0027 \u0027Self-Employment Profit,\u0027 and specific tax-related calculations, which are indicative of a tax-related document. The structure and terminology align with corporation tax or self-assessment tax documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document references \u0027Class 4 National Insurance Charge,\u0027 \u0027Self-Employment Profit,\u0027 and specific tax-related calculations, which are indicative of a tax-related document. The structure and terminology align with corporation tax or self-assessment tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 26,
"text_length": 362
}
],
"start_page": 25
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "financial_statement",
"end_page": 29,
"pages": [
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of transactions, invoices, payroll, loans, or property-related information.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKX2rv4IvMaSeyqIKP0KiHPl9aDs\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of transactions, invoices, payroll, loans, or property-related information.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504416, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=219, prompt_tokens=1394, total_tokens=1613, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of transactions, invoices, payroll, loans, or property-related information.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of transactions, invoices, payroll, loans, or property-related information.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.16941176470588235,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.23529411764705882,
"pattern_score": 0.0,
"total_score": 0.16941176470588235
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.15789473684210525,
"pattern_score": 0.1,
"total_score": 0.1616842105263158
},
"hp_agreement": {
"keyword_score": 0.038461538461538464,
"pattern_score": 0.0,
"total_score": 0.023076923076923078
},
"invoice": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.0,
"total_score": 0.09999999999999999
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "corporation_tax"
}
},
"page_number": 28,
"text_length": 1300
},
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes adjusted profit/loss calculations and references to a tax basis period, which are key indicators of a financial statement. There are no references to transactions, invoices, loans, mortgages, or other specific document types.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKX3qrvAQFu9tiAQsavUUZBOKRoe\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes adjusted profit/loss calculations and references to a tax basis period, which are key indicators of a financial statement. There are no references to transactions, invoices, loans, mortgages, or other specific document types.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504417, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=227, prompt_tokens=1291, total_tokens=1518, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_t"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes adjusted profit/loss calculations and references to a tax basis period, which are key indicators of a financial statement. There are no references to transactions, invoices, loans, mortgages, or other specific document types.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes adjusted profit/loss calculations and references to a tax basis period, which are key indicators of a financial statement. There are no references to transactions, invoices, loans, mortgages, or other specific document types.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.34357894736842104,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.21052631578947367,
"pattern_score": 0.4,
"total_score": 0.34357894736842104
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "financial_statement"
}
},
"page_number": 29,
"text_length": 874
}
],
"start_page": 28
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"completion_date": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"purchase_sale_price": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "property_statement",
"end_page": 31,
"pages": [
{
"confidence_score": 0.95,
"document_type": "property_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains details about UK property income, including property type, income from rents, expenses, and net profit. These are key indicators of a property statement. Additionally, the document mentions \u0027Property Details\u0027 and \u0027Rents From UK Property,\u0027 which are specific to property-related financial documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_sale_price\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKX5bd66ei8sTzT0cP1XXQ046Z6w\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"property_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains details about UK property income, including property type, income from rents, expenses, and net profit. These are key indicators of a property statement. Additionally, the document mentions \\\u0027Property Details\\\u0027 and \\\u0027Rents From UK Property,\\\u0027 which are specific to property-related financial documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null,\\n \"completion_date\": null,\\n \"property_address\": null,\\n \"purchase_sale_price\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504419, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=242, prompt_tokens=1130, total_tokens=1372, completion_tokens"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"completion_date": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"purchase_sale_price": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains details about UK property income, including property type, income from rents, expenses, and net profit. These are key indicators of a property statement. Additionally, the document mentions \u0027Property Details\u0027 and \u0027Rents From UK Property,\u0027 which are specific to property-related financial documents.",
"raw_response": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains details about UK property income, including property type, income from rents, expenses, and net profit. These are key indicators of a property statement. Additionally, the document mentions \u0027Property Details\u0027 and \u0027Rents From UK Property,\u0027 which are specific to property-related financial documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_sale_price\": null\n }\n}\n```"
},
"type": "property_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03789473684210525,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "financial_statement"
}
},
"page_number": 30,
"text_length": 425
},
{
"confidence_score": 0.95,
"document_type": "property_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to property finance charges, property business, adjusted total income, and tax relief related to property. These are key indicators of a property statement. Additionally, terms like \u0027Property Finance Charges Relief,\u0027 \u0027UK Property,\u0027 and \u0027Total Relief Available\u0027 strongly suggest this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_price\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKX64EDoOblJoF2dKo3Ot6QFNtDV\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"property_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to property finance charges, property business, adjusted total income, and tax relief related to property. These are key indicators of a property statement. Additionally, terms like \\\u0027Property Finance Charges Relief,\\\u0027 \\\u0027UK Property,\\\u0027 and \\\u0027Total Relief Available\\\u0027 strongly suggest this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null,\\n \"completion_date\": null,\\n \"property_address\": null,\\n \"purchase_price\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504420, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=240, prompt_tokens=1291, total_tokens=1531, completio"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"completion_date": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"purchase_price": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to property finance charges, property business, adjusted total income, and tax relief related to property. These are key indicators of a property statement. Additionally, terms like \u0027Property Finance Charges Relief,\u0027 \u0027UK Property,\u0027 and \u0027Total Relief Available\u0027 strongly suggest this classification.",
"raw_response": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to property finance charges, property business, adjusted total income, and tax relief related to property. These are key indicators of a property statement. Additionally, terms like \u0027Property Finance Charges Relief,\u0027 \u0027UK Property,\u0027 and \u0027Total Relief Available\u0027 strongly suggest this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_price\": null\n }\n}\n```"
},
"type": "property_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.038461538461538464,
"pattern_score": 0.0,
"total_score": 0.023076923076923078
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 31,
"text_length": 990
}
],
"start_page": 30
}
],
"page_classifications": [
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a Tax Return for a specific year, and mentions of Self Assessment Online Service, IRmark number, and tax repayment claims. These are key indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWMmDnPajuE7grpXycO4On8OYk0\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a Tax Return for a specific year, and mentions of Self Assessment Online Service, IRmark number, and tax repayment claims. These are key indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504374, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=209, prompt_tokens=1383, total_tokens=1592, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(audio_tokens=0, cached_to"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HM Revenue \u0026 Customs, a Tax Return for a specific year, and mentions of Self Assessment Online Service, IRmark number, and tax repayment claims. These are key indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a Tax Return for a specific year, and mentions of Self Assessment Online Service, IRmark number, and tax repayment claims. These are key indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.13270588235294117,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.1,
"total_score": 0.13270588235294117
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "corporation_tax"
}
},
"page_number": 1,
"text_length": 1899
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return, schedule reports, rate band analysis, client payment advice, payments on account, and other tax-related terms. These are strong indicators of a corporation tax document, likely related to a tax return or computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWNuOli5vXtiy4ynfGwNrhIHYkU\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to a tax return, schedule reports, rate band analysis, client payment advice, payments on account, and other tax-related terms. These are strong indicators of a corporation tax document, likely related to a tax return or computation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504375, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=212, prompt_tokens=1047, total_tokens=1259, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetai"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to a tax return, schedule reports, rate band analysis, client payment advice, payments on account, and other tax-related terms. These are strong indicators of a corporation tax document, likely related to a tax return or computation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return, schedule reports, rate band analysis, client payment advice, payments on account, and other tax-related terms. These are strong indicators of a corporation tax document, likely related to a tax return or computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.038461538461538464,
"pattern_score": 0.0,
"total_score": 0.023076923076923078
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 2,
"text_length": 249
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax return, tax year, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. Additionally, it includes details about deadlines for filing tax returns and penalties for late submissions, which are specific to tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWPM0Ixg35qMRqOLaSLmh1SzSNZ\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax return, tax year, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. Additionally, it includes details about deadlines for filing tax returns and penalties for late submissions, which are specific to tax-related documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504377, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=224, prompt_tokens=1727, total_tokens=1951, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax return, tax year, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. Additionally, it includes details about deadlines for filing tax returns and penalties for late submissions, which are specific to tax-related documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax return, tax year, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. Additionally, it includes details about deadlines for filing tax returns and penalties for late submissions, which are specific to tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.1807058823529412,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.2,
"total_score": 0.1807058823529412
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 3,
"text_length": 2696
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to a tax return, tax reference number, taxable income, capital gains tax, self-employment income, UK property income, and other tax-related information. These are strong indicators of a corporation tax document or a tax return form.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWQS2sBU9FfcUD8XNWTMQ8LIFEw\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to a tax return, tax reference number, taxable income, capital gains tax, self-employment income, UK property income, and other tax-related information. These are strong indicators of a corporation tax document or a tax return form.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504378, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1730, total_tokens=1944, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDeta"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to a tax return, tax reference number, taxable income, capital gains tax, self-employment income, UK property income, and other tax-related information. These are strong indicators of a corporation tax document or a tax return form.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to a tax return, tax reference number, taxable income, capital gains tax, self-employment income, UK property income, and other tax-related information. These are strong indicators of a corporation tax document or a tax return form.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 4,
"text_length": 4065
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027SA 100\u0027. These are key indicators of a tax-related document, specifically a corporation tax return or related form. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return document, which falls under the corporation_tax category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWSwbOpBxxycfZazovwqIGslk6D\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Taxed UK interest\\\u0027, \\\u0027Untaxed UK interest\\\u0027, \\\u0027Foreign dividends\\\u0027, \\\u0027Taxable Incapacity Benefit\\\u0027, and \\\u0027SA 100\\\u0027. These are key indicators of a tax-related document, specifically a corporation tax return or related form. The presence of \\\u0027SA 100\\\u0027 further confirms that this is a self-assessment tax return document, which falls under the corporation_tax category.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504380, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=256, prompt_tokens=1635, total_tokens="
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027SA 100\u0027. These are key indicators of a tax-related document, specifically a corporation tax return or related form. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return document, which falls under the corporation_tax category.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027SA 100\u0027. These are key indicators of a tax-related document, specifically a corporation tax return or related form. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return document, which falls under the corporation_tax category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.15,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.125,
"total_score": 0.15
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.030303030303030304,
"total_score": 0.03681818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 5,
"text_length": 2229
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and other tax-related allowances, which are typically part of a corporation tax return or related documentation. The presence of a tax reference number and the mention of \u0027SA 100\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWT3zve4Bm6jtLYusCpQHzDNIJc\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and other tax-related allowances, which are typically part of a corporation tax return or related documentation. The presence of a tax reference number and the mention of \\\u0027SA 100\\\u0027 further supports this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504381, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=219, prompt_tokens=1598, total_tokens=1817, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_token"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax reliefs, pension contributions, charitable giving, and other tax-related allowances, which are typically part of a corporation tax return or related documentation. The presence of a tax reference number and the mention of \u0027SA 100\u0027 further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and other tax-related allowances, which are typically part of a corporation tax return or related documentation. The presence of a tax reference number and the mention of \u0027SA 100\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.030303030303030304,
"total_score": 0.03681818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 6,
"text_length": 2174
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWU39VxwWRCh2RPtZQkxPtgQ2Eh\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Student Loan and Postgraduate Loan repayments\\\u0027, \\\u0027High Income Child Benefit Charge\\\u0027, and \\\u0027Marriage Allowance\\\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \\\u0027SA 100\\\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504382, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usag"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11454545454545456,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.09090909090909091,
"total_score": 0.11454545454545456
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.030303030303030304,
"total_score": 0.06681818181818183
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "loan_statement"
}
},
"page_number": 7,
"text_length": 2181
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax refunds, and specific tax years (e.g., 2024-25, 2025-26). It also mentions terms like \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary,\u0027 which are indicative of a corporation tax document. Additionally, the document includes a client reference number and tax reference, which are typical identifiers in tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWW042fG8HN0NpTROfUb7Q24NNf\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax refunds, and specific tax years (e.g., 2024-25, 2025-26). It also mentions terms like \\\u0027Self Assessment Statement,\\\u0027 \\\u0027Tax Reference,\\\u0027 and \\\u0027Tax calculation summary,\\\u0027 which are indicative of a corporation tax document. Additionally, the document includes a client reference number and tax reference, which are typical identifiers in tax-related documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504384, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=248, prompt_tokens=1673, total_tokens=1921, completion_tokens_detai"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax returns, tax calculations, tax refunds, and specific tax years (e.g., 2024-25, 2025-26). It also mentions terms like \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary,\u0027 which are indicative of a corporation tax document. Additionally, the document includes a client reference number and tax reference, which are typical identifiers in tax-related documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax refunds, and specific tax years (e.g., 2024-25, 2025-26). It also mentions terms like \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary,\u0027 which are indicative of a corporation tax document. Additionally, the document includes a client reference number and tax reference, which are typical identifiers in tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.1392857142857143,
"features": {
"bank_statement": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.0,
"total_score": 0.10588235294117647
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.14285714285714285,
"total_score": 0.1392857142857143
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.030303030303030304,
"total_score": 0.05181818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 8,
"text_length": 2552
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax adviser, a tax reference number, and mentions \u0027Income from property\u0027 and \u0027SA 100\u0027, which are indicative of tax-related documentation. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Client Ref\u0027, and \u0027SA 100\u0027 strongly suggests this is related to corporation tax or self-assessment tax returns.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWYjJe69dwb6PhFCWwpZJ2xCmr8\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to a tax adviser, a tax reference number, and mentions \\\u0027Income from property\\\u0027 and \\\u0027SA 100\\\u0027, which are indicative of tax-related documentation. Additionally, the presence of terms like \\\u0027Tax Reference\\\u0027, \\\u0027Client Ref\\\u0027, and \\\u0027SA 100\\\u0027 strongly suggests this is related to corporation tax or self-assessment tax returns.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504386, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=234, prompt_tokens=1187, total_tokens=1421, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, r"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to a tax adviser, a tax reference number, and mentions \u0027Income from property\u0027 and \u0027SA 100\u0027, which are indicative of tax-related documentation. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Client Ref\u0027, and \u0027SA 100\u0027 strongly suggests this is related to corporation tax or self-assessment tax returns.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax adviser, a tax reference number, and mentions \u0027Income from property\u0027 and \u0027SA 100\u0027, which are indicative of tax-related documentation. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Client Ref\u0027, and \u0027SA 100\u0027 strongly suggests this is related to corporation tax or self-assessment tax returns.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 9,
"text_length": 635
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return (SA100), tax reference number, and a declaration section, which are key indicators of a corporation tax document. Additionally, it includes instructions for filling out and signing the tax return, which aligns with the characteristics of corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWaQ53wUyuAXipTGFFEeD9AevV8\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to a tax return (SA100), tax reference number, and a declaration section, which are key indicators of a corporation tax document. Additionally, it includes instructions for filling out and signing the tax return, which aligns with the characteristics of corporation tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504388, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=218, prompt_tokens=1262, total_tokens=1480, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tok"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to a tax return (SA100), tax reference number, and a declaration section, which are key indicators of a corporation tax document. Additionally, it includes instructions for filling out and signing the tax return, which aligns with the characteristics of corporation tax documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return (SA100), tax reference number, and a declaration section, which are key indicators of a corporation tax document. Additionally, it includes instructions for filling out and signing the tax return, which aligns with the characteristics of corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.0,
"total_score": 0.045
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 10,
"text_length": 1070
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, Unique Taxpayer Reference (UTR), business details, turnover, and other income details, which are key indicators of a corporation tax document. Additionally, the form is labeled as \u0027SA103F,\u0027 which is a self-employment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWbUjNSEKMQ7P7GxrzNz9Y26nzb\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax year, Unique Taxpayer Reference (UTR), business details, turnover, and other income details, which are key indicators of a corporation tax document. Additionally, the form is labeled as \\\u0027SA103F,\\\u0027 which is a self-employment tax return form used in the UK.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504389, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=229, prompt_tokens=1570, total_tokens=1799, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), p"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax year, Unique Taxpayer Reference (UTR), business details, turnover, and other income details, which are key indicators of a corporation tax document. Additionally, the form is labeled as \u0027SA103F,\u0027 which is a self-employment tax return form used in the UK.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, Unique Taxpayer Reference (UTR), business details, turnover, and other income details, which are key indicators of a corporation tax document. Additionally, the form is labeled as \u0027SA103F,\u0027 which is a self-employment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.21741176470588236,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.23529411764705882,
"pattern_score": 0.1,
"total_score": 0.21741176470588236
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.15789473684210525,
"pattern_score": 0.0,
"total_score": 0.11368421052631578
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.0,
"total_score": 0.13
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 11,
"text_length": 1871
},
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains detailed financial information related to business expenses, including categories such as wages, salaries, travel expenses, rent, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWcpe1HU5hkrHzH93RahS30vFmZ\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains detailed financial information related to business expenses, including categories such as wages, salaries, travel expenses, rent, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504390, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=229, prompt_tokens=1585, total_tokens=1814, completion_tokens_details=CompletionToken"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains detailed financial information related to business expenses, including categories such as wages, salaries, travel expenses, rent, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains detailed financial information related to business expenses, including categories such as wages, salaries, travel expenses, rent, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.03529411764705882
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.0,
"total_score": 0.04
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 12,
"text_length": 1488
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related terms such as \u0027Tax Reference\u0027, \u0027Net profit or loss\u0027, \u0027Tax allowances for vehicles and equipment\u0027, \u0027capital allowances\u0027, and \u0027taxable profit or loss\u0027. Additionally, it includes specific tax-related calculations and references to HMRC forms (e.g., SA103F). These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWePuli9M2fjEwYtcYmIbpwwhn0\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related terms such as \\\u0027Tax Reference\\\u0027, \\\u0027Net profit or loss\\\u0027, \\\u0027Tax allowances for vehicles and equipment\\\u0027, \\\u0027capital allowances\\\u0027, and \\\u0027taxable profit or loss\\\u0027. Additionally, it includes specific tax-related calculations and references to HMRC forms (e.g., SA103F). These are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504392, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=236, prompt_tokens=1724, total_tokens=1960, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0,"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related terms such as \u0027Tax Reference\u0027, \u0027Net profit or loss\u0027, \u0027Tax allowances for vehicles and equipment\u0027, \u0027capital allowances\u0027, and \u0027taxable profit or loss\u0027. Additionally, it includes specific tax-related calculations and references to HMRC forms (e.g., SA103F). These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related terms such as \u0027Tax Reference\u0027, \u0027Net profit or loss\u0027, \u0027Tax allowances for vehicles and equipment\u0027, \u0027capital allowances\u0027, and \u0027taxable profit or loss\u0027. Additionally, it includes specific tax-related calculations and references to HMRC forms (e.g., SA103F). These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 13,
"text_length": 2548
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWgAUnNXIwHfliac8ATt1aC0Gka\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \\\u0027Self-employment (full) notes,\\\u0027 \\\u0027taxable profits,\\\u0027 and \\\u0027losses,\\\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504394, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=253, prompt_tokens=1729, total_tokens=1982, co"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.08470588235294117,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.08470588235294117
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 14,
"text_length": 2609
},
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital of a business. Specific terms such as \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 and \u0027Net business assets\u0027 are present, which are key indicators of a financial statement. There is no evidence of transactions, invoices, payroll, or other document types listed in the categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWhkkYAd0tc4PYchrKpvd8s5xUt\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital of a business. Specific terms such as \\\u0027Balance sheet,\\\u0027 \\\u0027Assets,\\\u0027 \\\u0027Liabilities,\\\u0027 and \\\u0027Net business assets\\\u0027 are present, which are key indicators of a financial statement. There is no evidence of transactions, invoices, payroll, or other document types listed in the categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504395, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=235, prompt_tokens=1539, total_tokens=1774, completion_tokens_details=CompletionTokensDetails(accepted_p"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a balance sheet with detailed information about assets, liabilities, and capital of a business. Specific terms such as \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 and \u0027Net business assets\u0027 are present, which are key indicators of a financial statement. There is no evidence of transactions, invoices, payroll, or other document types listed in the categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital of a business. Specific terms such as \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 and \u0027Net business assets\u0027 are present, which are key indicators of a financial statement. There is no evidence of transactions, invoices, payroll, or other document types listed in the categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.2374736842105263,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.2631578947368421,
"pattern_score": 0.1,
"total_score": 0.2374736842105263
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 15,
"text_length": 1430
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and specific details about Class 2 and Class 4 NICs. It also includes a client reference and mentions self-employment profits, which are indicative of a tax-related document. The format and content align with HMRC tax documentation, specifically related to self-assessment forms.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWjA5AdDrEmvlMjbDNTbx5Ga7rA\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and specific details about Class 2 and Class 4 NICs. It also includes a client reference and mentions self-employment profits, which are indicative of a tax-related document. The format and content align with HMRC tax documentation, specifically related to self-assessment forms.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504397, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=240, prompt_tokens=1210, total_tokens=1450, completion_tokens_details=Completion"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and specific details about Class 2 and Class 4 NICs. It also includes a client reference and mentions self-employment profits, which are indicative of a tax-related document. The format and content align with HMRC tax documentation, specifically related to self-assessment forms.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and specific details about Class 2 and Class 4 NICs. It also includes a client reference and mentions self-employment profits, which are indicative of a tax-related document. The format and content align with HMRC tax documentation, specifically related to self-assessment forms.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 16,
"text_length": 764
},
{
"confidence_score": 0.95,
"document_type": "property_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to UK property details, rental income, allowable expenses, and other property-related financial information. It also includes terms like \u0027property income,\u0027 \u0027rented out,\u0027 and \u0027furnished holiday lettings,\u0027 which are key indicators of a property statement. Additionally, the document is labeled as \u0027SA 105,\u0027 which is a supplementary page for property income in UK tax returns.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_price\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWkTzT1AIqCyt9UY9VmCM2dRtFZ\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"property_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to UK property details, rental income, allowable expenses, and other property-related financial information. It also includes terms like \\\u0027property income,\\\u0027 \\\u0027rented out,\\\u0027 and \\\u0027furnished holiday lettings,\\\u0027 which are key indicators of a property statement. Additionally, the document is labeled as \\\u0027SA 105,\\\u0027 which is a supplementary page for property income in UK tax returns.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null,\\n \"completion_date\": null,\\n \"property_address\": null,\\n \"purchase_price\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504398, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=Completio"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"completion_date": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"purchase_price": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to UK property details, rental income, allowable expenses, and other property-related financial information. It also includes terms like \u0027property income,\u0027 \u0027rented out,\u0027 and \u0027furnished holiday lettings,\u0027 which are key indicators of a property statement. Additionally, the document is labeled as \u0027SA 105,\u0027 which is a supplementary page for property income in UK tax returns.",
"raw_response": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to UK property details, rental income, allowable expenses, and other property-related financial information. It also includes terms like \u0027property income,\u0027 \u0027rented out,\u0027 and \u0027furnished holiday lettings,\u0027 which are key indicators of a property statement. Additionally, the document is labeled as \u0027SA 105,\u0027 which is a supplementary page for property income in UK tax returns.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_price\": null\n }\n}\n```"
},
"type": "property_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.21741176470588236,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.23529411764705882,
"pattern_score": 0.1,
"total_score": 0.21741176470588236
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.0,
"total_score": 0.06
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.0,
"total_score": 0.045
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 17,
"text_length": 2125
},
{
"confidence_score": 0.9,
"document_type": "property_statement",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to property income, property expenses, residential property finance costs, and property-related allowances. It also includes a client reference and tax reference, which are common in property-related financial documents. The presence of terms like \u0027Property income\u0027, \u0027Property expenses\u0027, and \u0027Residential property finance costs\u0027 strongly indicates that this is a property statement.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_sale_price\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWme6wSu95XTFKw81I7Zavyty4j\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"property_statement\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to property income, property expenses, residential property finance costs, and property-related allowances. It also includes a client reference and tax reference, which are common in property-related financial documents. The presence of terms like \\\u0027Property income\\\u0027, \\\u0027Property expenses\\\u0027, and \\\u0027Residential property finance costs\\\u0027 strongly indicates that this is a property statement.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null,\\n \"completion_date\": null,\\n \"property_address\": null,\\n \"purchase_sale_price\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504400, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"completion_date": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"purchase_sale_price": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to property income, property expenses, residential property finance costs, and property-related allowances. It also includes a client reference and tax reference, which are common in property-related financial documents. The presence of terms like \u0027Property income\u0027, \u0027Property expenses\u0027, and \u0027Residential property finance costs\u0027 strongly indicates that this is a property statement.",
"raw_response": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to property income, property expenses, residential property finance costs, and property-related allowances. It also includes a client reference and tax reference, which are common in property-related financial documents. The presence of terms like \u0027Property income\u0027, \u0027Property expenses\u0027, and \u0027Residential property finance costs\u0027 strongly indicates that this is a property statement.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_sale_price\": null\n }\n}\n```"
},
"type": "property_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 18,
"text_length": 2119
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax calculation summary, Self Assessment, and specific tax-related terms such as \u0027Tax year\u0027, \u0027Unique Taxpayer Reference (UTR)\u0027, \u0027Class 2 NICs\u0027, \u0027Class 4 NICs\u0027, \u0027Capital Gains Tax\u0027, and \u0027Payments on account\u0027. These are clear indicators of a tax-related document, specifically a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWod4NzhIvZhmIaE5rXX0MFHiFl\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax calculation summary, Self Assessment, and specific tax-related terms such as \\\u0027Tax year\\\u0027, \\\u0027Unique Taxpayer Reference (UTR)\\\u0027, \\\u0027Class 2 NICs\\\u0027, \\\u0027Class 4 NICs\\\u0027, \\\u0027Capital Gains Tax\\\u0027, and \\\u0027Payments on account\\\u0027. These are clear indicators of a tax-related document, specifically a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504402, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=240, prompt_tokens=1689, total_tokens=1929, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax calculation summary, Self Assessment, and specific tax-related terms such as \u0027Tax year\u0027, \u0027Unique Taxpayer Reference (UTR)\u0027, \u0027Class 2 NICs\u0027, \u0027Class 4 NICs\u0027, \u0027Capital Gains Tax\u0027, and \u0027Payments on account\u0027. These are clear indicators of a tax-related document, specifically a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax calculation summary, Self Assessment, and specific tax-related terms such as \u0027Tax year\u0027, \u0027Unique Taxpayer Reference (UTR)\u0027, \u0027Class 2 NICs\u0027, \u0027Class 4 NICs\u0027, \u0027Capital Gains Tax\u0027, and \u0027Payments on account\u0027. These are clear indicators of a tax-related document, specifically a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.17505882352941174,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.1,
"total_score": 0.17505882352941174
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.09090909090909091,
"total_score": 0.11454545454545456
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.06060606060606061,
"total_score": 0.08863636363636364
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 19,
"text_length": 2286
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027tax due,\u0027 \u0027repayment,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWqZnFJjJjKYBDQCS06KiHPLKZA\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \\\u0027tax due,\\\u0027 \\\u0027repayment,\\\u0027 and \\\u0027self-assessment statement of account.\\\u0027 These are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504404, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=224, prompt_tokens=1326, total_tokens=1550, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptToken"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027tax due,\u0027 \u0027repayment,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027tax due,\u0027 \u0027repayment,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.07058823529411765,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "bank_statement"
}
},
"page_number": 20,
"text_length": 1359
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains tax-related information, including a computation summary, taxable income, tax rates, and National Insurance Contributions (NIC). It also references a tax year (2024/25) and includes a tax reference number, which are strong indicators of a tax-related document. The structure and content align closely with a tax return or computation document, which falls under the \u0027corporation_tax\u0027 category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWrZuhfzgpjLH8oZFjFHSmvS6Nx\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains tax-related information, including a computation summary, taxable income, tax rates, and National Insurance Contributions (NIC). It also references a tax year (2024/25) and includes a tax reference number, which are strong indicators of a tax-related document. The structure and content align closely with a tax return or computation document, which falls under the \\\u0027corporation_tax\\\u0027 category.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504405, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=242, prompt_tokens=1327, total_tokens=1569, completion_tokens_details=Completi"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains tax-related information, including a computation summary, taxable income, tax rates, and National Insurance Contributions (NIC). It also references a tax year (2024/25) and includes a tax reference number, which are strong indicators of a tax-related document. The structure and content align closely with a tax return or computation document, which falls under the \u0027corporation_tax\u0027 category.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains tax-related information, including a computation summary, taxable income, tax rates, and National Insurance Contributions (NIC). It also references a tax year (2024/25) and includes a tax reference number, which are strong indicators of a tax-related document. The structure and content align closely with a tax return or computation document, which falls under the \u0027corporation_tax\u0027 category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.038461538461538464,
"pattern_score": 0.0,
"total_score": 0.023076923076923078
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.030303030303030304,
"total_score": 0.08181818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 21,
"text_length": 1026
},
{
"confidence_score": 0.85,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWtaSM0Enrox0EyJ4mju2XXqS1Q\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504407, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1970, total_tokens=2184, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prom"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 22,
"text_length": 1715
},
{
"confidence_score": 0.8,
"document_type": "unknown",
"features": {
"llm_based": {
"confidence": 0.8,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.8,\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, and other financial metrics, which are indicative of a financial statement. There is no evidence of transactions, invoices, payroll, or other specific document types listed in the categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWubrlH4lToea35um14VuZzC1Ua\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.8,\\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, and other financial metrics, which are indicative of a financial statement. There is no evidence of transactions, invoices, payroll, or other specific document types listed in the categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504408, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1114, total_tokens=1328, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_toke"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to capital gains tax (CGT), business asset disposal relief gains, and other financial metrics, which are indicative of a financial statement. There is no evidence of transactions, invoices, payroll, or other specific document types listed in the categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.8,\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, and other financial metrics, which are indicative of a financial statement. There is no evidence of transactions, invoices, payroll, or other specific document types listed in the categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "uncertain",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 23,
"text_length": 395
},
{
"confidence_score": 0.95,
"document_type": "p32_payroll",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and \u0027Amounts Due To Be Paid Via Instalments,\u0027 which are indicative of HMRC P32 Employer Payment Records. Additionally, it includes specific payment dates and amounts, which align with tax obligations and employer payment records.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWv86pprGbOWXmHXaFaLHZhmEuU\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"p32_payroll\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to \\\u0027Client Payment Advice,\\\u0027 \\\u0027Payments Required,\\\u0027 \\\u0027Payment Dates,\\\u0027 and \\\u0027Amounts Due To Be Paid Via Instalments,\\\u0027 which are indicative of HMRC P32 Employer Payment Records. Additionally, it includes specific payment dates and amounts, which align with tax obligations and employer payment records.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504409, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=225, prompt_tokens=1135, total_tokens=1360, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rej"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and \u0027Amounts Due To Be Paid Via Instalments,\u0027 which are indicative of HMRC P32 Employer Payment Records. Additionally, it includes specific payment dates and amounts, which align with tax obligations and employer payment records.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and \u0027Amounts Due To Be Paid Via Instalments,\u0027 which are indicative of HMRC P32 Employer Payment Records. Additionally, it includes specific payment dates and amounts, which align with tax obligations and employer payment records.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "p32_payroll"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 24,
"text_length": 397
},
{
"confidence_score": 0.85,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of a tax computation or return. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting. These indicators suggest the document is related to corporation tax or a similar tax computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWxzYW1h4ckx0zuI5yoX3yj6eUx\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document references tax-related terms such as \\\u0027Tax Due After Notional Taxes Paid,\\\u0027 \\\u0027Class 4 NIC Due,\\\u0027 and \\\u0027Payments on Account,\\\u0027 which are indicative of a tax computation or return. The mention of \\\u0027Year ended 5 April 2025\\\u0027 aligns with tax year reporting. These indicators suggest the document is related to corporation tax or a similar tax computation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504411, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=238, prompt_tokens=1109, total_tokens=1347, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_t"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of a tax computation or return. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting. These indicators suggest the document is related to corporation tax or a similar tax computation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of a tax computation or return. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting. These indicators suggest the document is related to corporation tax or a similar tax computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 25,
"text_length": 374
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document references \u0027Class 4 National Insurance Charge,\u0027 \u0027Self-Employment Profit,\u0027 and specific tax-related calculations, which are indicative of a tax-related document. The structure and terminology align with corporation tax or self-assessment tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKWz5JTXzgHRKHdmIxSq4HzFwhzG\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document references \\\u0027Class 4 National Insurance Charge,\\\u0027 \\\u0027Self-Employment Profit,\\\u0027 and specific tax-related calculations, which are indicative of a tax-related document. The structure and terminology align with corporation tax or self-assessment tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504413, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=210, prompt_tokens=1121, total_tokens=1331, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDet"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document references \u0027Class 4 National Insurance Charge,\u0027 \u0027Self-Employment Profit,\u0027 and specific tax-related calculations, which are indicative of a tax-related document. The structure and terminology align with corporation tax or self-assessment tax documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document references \u0027Class 4 National Insurance Charge,\u0027 \u0027Self-Employment Profit,\u0027 and specific tax-related calculations, which are indicative of a tax-related document. The structure and terminology align with corporation tax or self-assessment tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 26,
"text_length": 362
},
{
"confidence_score": 0.8,
"document_type": "unknown",
"features": {
"llm_based": {
"confidence": 0.8,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.8,\n \"reasoning\": \"The document mentions \u0027Year ended 5 April 2025\u0027 and \u0027Personal Allowances,\u0027 which are indicative of a financial summary or statement. However, it lacks specific indicators of other categories such as transactions, invoices, or loan details. The absence of transactional data, account numbers, or property-related information further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKX0uDxbBj7H6LvfX2QUHBCLHphN\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.8,\\n \"reasoning\": \"The document mentions \\\u0027Year ended 5 April 2025\\\u0027 and \\\u0027Personal Allowances,\\\u0027 which are indicative of a financial summary or statement. However, it lacks specific indicators of other categories such as transactions, invoices, or loan details. The absence of transactional data, account numbers, or property-related information further supports this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504414, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=228, prompt_tokens=1055, total_tokens=1283, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, aud"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document mentions \u0027Year ended 5 April 2025\u0027 and \u0027Personal Allowances,\u0027 which are indicative of a financial summary or statement. However, it lacks specific indicators of other categories such as transactions, invoices, or loan details. The absence of transactional data, account numbers, or property-related information further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.8,\n \"reasoning\": \"The document mentions \u0027Year ended 5 April 2025\u0027 and \u0027Personal Allowances,\u0027 which are indicative of a financial summary or statement. However, it lacks specific indicators of other categories such as transactions, invoices, or loan details. The absence of transactional data, account numbers, or property-related information further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "uncertain",
"pattern_based": {
"confidence": 0.0,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "bank_statement"
}
},
"page_number": 27,
"text_length": 179
},
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of transactions, invoices, payroll, loans, or property-related information.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKX2rv4IvMaSeyqIKP0KiHPl9aDs\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of transactions, invoices, payroll, loans, or property-related information.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504416, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=219, prompt_tokens=1394, total_tokens=1613, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of transactions, invoices, payroll, loans, or property-related information.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of transactions, invoices, payroll, loans, or property-related information.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.16941176470588235,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.23529411764705882,
"pattern_score": 0.0,
"total_score": 0.16941176470588235
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.15789473684210525,
"pattern_score": 0.1,
"total_score": 0.1616842105263158
},
"hp_agreement": {
"keyword_score": 0.038461538461538464,
"pattern_score": 0.0,
"total_score": 0.023076923076923078
},
"invoice": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.0,
"total_score": 0.09999999999999999
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "corporation_tax"
}
},
"page_number": 28,
"text_length": 1300
},
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes adjusted profit/loss calculations and references to a tax basis period, which are key indicators of a financial statement. There are no references to transactions, invoices, loans, mortgages, or other specific document types.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKX3qrvAQFu9tiAQsavUUZBOKRoe\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes adjusted profit/loss calculations and references to a tax basis period, which are key indicators of a financial statement. There are no references to transactions, invoices, loans, mortgages, or other specific document types.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504417, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=227, prompt_tokens=1291, total_tokens=1518, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_t"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes adjusted profit/loss calculations and references to a tax basis period, which are key indicators of a financial statement. There are no references to transactions, invoices, loans, mortgages, or other specific document types.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes adjusted profit/loss calculations and references to a tax basis period, which are key indicators of a financial statement. There are no references to transactions, invoices, loans, mortgages, or other specific document types.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.34357894736842104,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.21052631578947367,
"pattern_score": 0.4,
"total_score": 0.34357894736842104
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "financial_statement"
}
},
"page_number": 29,
"text_length": 874
},
{
"confidence_score": 0.95,
"document_type": "property_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains details about UK property income, including property type, income from rents, expenses, and net profit. These are key indicators of a property statement. Additionally, the document mentions \u0027Property Details\u0027 and \u0027Rents From UK Property,\u0027 which are specific to property-related financial documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_sale_price\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKX5bd66ei8sTzT0cP1XXQ046Z6w\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"property_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains details about UK property income, including property type, income from rents, expenses, and net profit. These are key indicators of a property statement. Additionally, the document mentions \\\u0027Property Details\\\u0027 and \\\u0027Rents From UK Property,\\\u0027 which are specific to property-related financial documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null,\\n \"completion_date\": null,\\n \"property_address\": null,\\n \"purchase_sale_price\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504419, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=242, prompt_tokens=1130, total_tokens=1372, completion_tokens"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"completion_date": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"purchase_sale_price": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains details about UK property income, including property type, income from rents, expenses, and net profit. These are key indicators of a property statement. Additionally, the document mentions \u0027Property Details\u0027 and \u0027Rents From UK Property,\u0027 which are specific to property-related financial documents.",
"raw_response": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains details about UK property income, including property type, income from rents, expenses, and net profit. These are key indicators of a property statement. Additionally, the document mentions \u0027Property Details\u0027 and \u0027Rents From UK Property,\u0027 which are specific to property-related financial documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_sale_price\": null\n }\n}\n```"
},
"type": "property_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03789473684210525,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "financial_statement"
}
},
"page_number": 30,
"text_length": 425
},
{
"confidence_score": 0.95,
"document_type": "property_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to property finance charges, property business, adjusted total income, and tax relief related to property. These are key indicators of a property statement. Additionally, terms like \u0027Property Finance Charges Relief,\u0027 \u0027UK Property,\u0027 and \u0027Total Relief Available\u0027 strongly suggest this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_price\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKX64EDoOblJoF2dKo3Ot6QFNtDV\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"property_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to property finance charges, property business, adjusted total income, and tax relief related to property. These are key indicators of a property statement. Additionally, terms like \\\u0027Property Finance Charges Relief,\\\u0027 \\\u0027UK Property,\\\u0027 and \\\u0027Total Relief Available\\\u0027 strongly suggest this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null,\\n \"completion_date\": null,\\n \"property_address\": null,\\n \"purchase_price\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504420, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=240, prompt_tokens=1291, total_tokens=1531, completio"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"completion_date": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"purchase_price": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to property finance charges, property business, adjusted total income, and tax relief related to property. These are key indicators of a property statement. Additionally, terms like \u0027Property Finance Charges Relief,\u0027 \u0027UK Property,\u0027 and \u0027Total Relief Available\u0027 strongly suggest this classification.",
"raw_response": "```json\n{\n \"document_type\": \"property_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to property finance charges, property business, adjusted total income, and tax relief related to property. These are key indicators of a property statement. Additionally, terms like \u0027Property Finance Charges Relief,\u0027 \u0027UK Property,\u0027 and \u0027Total Relief Available\u0027 strongly suggest this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null,\n \"completion_date\": null,\n \"property_address\": null,\n \"purchase_price\": null\n }\n}\n```"
},
"type": "property_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.038461538461538464,
"pattern_score": 0.0,
"total_score": 0.023076923076923078
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 31,
"text_length": 990
}
],
"processing_timestamp": "2026-05-11T13:00:22.411440"
}