| Original Filename: | WIN002_Winter_D_M_SA100_24-25_1777294387.pdf |
| Upload Date: | April 27, 2026 at 12:55 PM |
| Total Pages: | 28 |
| Processing Status: | Completed |
| Total Documents: | 12 |
| Corporation Taxs: | 7 |
| Financial Statements: | 4 |
| P32 Payrolls: | 1 |
Corporation Tax
95%Corporation Tax
90%Corporation Tax
95%Corporation Tax
90%Corporation Tax
90%Corporation Tax
90%Corporation Tax
85%Corporation Tax
90%Corporation Tax
85%Corporation Tax
85%Corporation Tax
95%Financial Statement
90%Corporation Tax
95%Corporation Tax
95%Financial Statement
95%Corporation Tax
90%Corporation Tax
95%Corporation Tax
95%Corporation Tax
95%Financial Statement
85%Financial Statement
85%P32 Payroll
90%Corporation Tax
85%Corporation Tax
85%Unknown
80%Financial Statement
95%Financial Statement
95%Unknown
80%| Document Type | Pages | Confidence | Processing Status | Created | Actions |
|---|---|---|---|---|---|
| Corporation Tax | 1-2 (2 pages) | 95% | Processed | 2026-04-27 12:55 | No view available |
| Corporation Tax | 3-11 (9 pages) | 95% | Processed | 2026-04-27 12:55 | No view available |
| Financial Statement | 12 (1 page) | 90% | Processed | 2026-04-27 12:55 | No view available |
| Corporation Tax | 13-14 (2 pages) | 95% | Processed | 2026-04-27 12:55 | No view available |
| Financial Statement | 15 (1 page) | 95% | Processed | 2026-04-27 12:56 | No view available |
| Corporation Tax | 16 (1 page) | 95% | Processed | 2026-04-27 12:56 | No view available |
| Corporation Tax | 17 (1 page) | 95% | Processed | 2026-04-27 12:56 | No view available |
| Corporation Tax | 18-19 (2 pages) | 90% | Processed | 2026-04-27 12:56 | No view available |
| Financial Statement | 20-21 (2 pages) | 85% | Processed | 2026-04-27 12:56 | No view available |
| P32 Payroll | 22 (1 page) | 90% | Processed | 2026-04-27 12:56 | No view available |
| Corporation Tax | 23-24 (2 pages) | 85% | Processed | 2026-04-27 12:56 | No view available |
| Financial Statement | 26-27 (2 pages) | 95% | Processed | 2026-04-27 12:56 | No view available |
{
"document_groups": [
{
"avg_confidence": 0.9,
"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 references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, it includes details about tax return submission and approval, which aligns with corporation tax documentation.\",\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-DZFmEoZMR7aG2qef5lUzPGRLoZ4sc\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 references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, it includes details about tax return submission and approval, which aligns with corporation tax documentation.\",\\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=1777294498, 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=1385, total_tokens=1609, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_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 references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, it includes details about tax return submission and approval, which aligns with corporation tax documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, it includes details about tax return submission and approval, which aligns with corporation tax documentation.\",\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": 1900
},
{
"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 mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are indicative of a corporation tax document or a tax return filing.\",\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-DZFmGJXel3Jkho0CfQU5iqym9hrxm\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 mentions \\\u00272024/25 Tax Return\\\u0027 and includes references to tax-related schedules such as \\\u0027Rate Band Analysis\\\u0027, \\\u0027Payments on Account\\\u0027, and \\\u0027Class 4 National Insurance Charge\\\u0027. These are indicative of a corporation tax document or a tax return filing.\",\\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=1777294500, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=216, prompt_tokens=1040, total_tokens=1256, 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 mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are indicative of a corporation tax document or a tax return filing.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are indicative of a corporation tax document or a tax return filing.\",\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": 2,
"text_length": 213
},
{
"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, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. It also includes details about tax deadlines, penalties, and instructions for filing a tax return, which are specific to corporation tax filings.\",\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-DZFmH0uKZsNJQ7Gk3df6URWKdNdZe\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, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. It also includes details about tax deadlines, penalties, and instructions for filing a tax return, which are specific to corporation tax filings.\",\\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=1777294501, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=223, prompt_tokens=1737, total_tokens=1960, 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 HMRC, tax return, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. It also includes details about tax deadlines, penalties, and instructions for filing a tax return, which are specific to corporation tax filings.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax return, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. It also includes details about tax deadlines, penalties, and instructions for filing a tax return, which are specific to corporation tax filings.\",\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": 2711
},
{
"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 returns, taxable income, supplementary pages for employment, self-employment, capital gains, and UK property income. These are key indicators of a corporation tax-related document, specifically a tax return or related 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-DZFmJmlf4eDtBQwzgv8ldO2zU5kcD\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 returns, taxable income, supplementary pages for employment, self-employment, capital gains, and UK property income. These are key indicators of a corporation tax-related document, specifically a tax return or related 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=1777294503, 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=1731, total_tokens=1941, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails("
},
"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, taxable income, supplementary pages for employment, self-employment, capital gains, and UK property income. These are key indicators of a corporation tax-related document, specifically a tax return or related form.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax returns, taxable income, supplementary pages for employment, self-employment, capital gains, and UK property income. These are key indicators of a corporation tax-related document, specifically a tax return or related 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": 0.9,
"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-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Dividends\u0027, \u0027State Pension\u0027, and \u0027Taxable Incapacity Benefit\u0027. These are indicative of a tax return or corporation tax document. Additionally, the presence of \u0027SA100\u0027 suggests it is related to a self-assessment tax return form, which aligns with corporation tax filings.\",\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-DZFmLOxdUYYph76n63QsiqqRFRYPI\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-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Taxed UK interest\\\u0027, \\\u0027Untaxed UK interest\\\u0027, \\\u0027Dividends\\\u0027, \\\u0027State Pension\\\u0027, and \\\u0027Taxable Incapacity Benefit\\\u0027. These are indicative of a tax return or corporation tax document. Additionally, the presence of \\\u0027SA100\\\u0027 suggests it is related to a self-assessment tax return form, which aligns with corporation tax filings.\",\\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=1777294505, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=249, prompt_tokens=1631, total_tokens=1880, completion_tokens_details=CompletionT"
},
"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, \u0027Dividends\u0027, \u0027State Pension\u0027, and \u0027Taxable Incapacity Benefit\u0027. These are indicative of a tax return or corporation tax document. Additionally, the presence of \u0027SA100\u0027 suggests it is related to a self-assessment tax return form, which aligns with corporation tax filings.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Dividends\u0027, \u0027State Pension\u0027, and \u0027Taxable Incapacity Benefit\u0027. These are indicative of a tax return or corporation tax document. Additionally, the presence of \u0027SA100\u0027 suggests it is related to a self-assessment tax return form, which aligns with corporation tax filings.\",\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": 2225
},
{
"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 allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\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-DZFmMpN905G7ireyJXvSJPQCMLMAl\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 allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \\\u0027Tax Reference\\\u0027, \\\u0027Gift Aid\\\u0027, and \\\u0027SA 100\\\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\\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=1777294506, 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=1597, total_tokens=1824, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning"
},
"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 allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.",
"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 allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\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": 2173
},
{
"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 contains references to tax-related information, such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, and \u0027High Income Child Benefit Charge\u0027. These are indicative of a tax-related document, specifically related to personal or corporate tax filings. The presence of terms like \u0027SA 100\u0027 and detailed tax-related sections 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-DZFmPyMhOwLR3unO5JIEmJT7MDL0p\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 contains references to tax-related information, such as \\\u0027Tax Reference\\\u0027, \\\u0027Student Loan and Postgraduate Loan repayments\\\u0027, and \\\u0027High Income Child Benefit Charge\\\u0027. These are indicative of a tax-related document, specifically related to personal or corporate tax filings. The presence of terms like \\\u0027SA 100\\\u0027 and detailed tax-related sections 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=1777294509, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=233, prompt_tokens=1537, total_tokens=1770, completion_tokens_details=CompletionTokensDetails(acce"
},
"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, and \u0027High Income Child Benefit Charge\u0027. These are indicative of a tax-related document, specifically related to personal or corporate tax filings. The presence of terms like \u0027SA 100\u0027 and detailed tax-related sections further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to tax-related information, such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, and \u0027High Income Child Benefit Charge\u0027. These are indicative of a tax-related document, specifically related to personal or corporate tax filings. The presence of terms like \u0027SA 100\u0027 and detailed tax-related sections 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": 0.11454545454545456,
"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": 2182
},
{
"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 returns, tax calculations, tax years, and HMRC-related processes, which are key indicators of corporation tax documents. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax calculation summary,\u0027 and \u0027Tax refunded or set off\u0027 further confirm 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-DZFmRIpadLO4DCmlHymiJXyoplcjE\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 returns, tax calculations, tax years, and HMRC-related processes, which are key indicators of corporation tax documents. Specific mentions of \\\u0027Self Assessment Statement,\\\u0027 \\\u0027Tax calculation summary,\\\u0027 and \\\u0027Tax refunded or set off\\\u0027 further confirm 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=1777294511, 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=1676, total_tokens=1895, 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 returns, tax calculations, tax years, and HMRC-related processes, which are key indicators of corporation tax documents. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax calculation summary,\u0027 and \u0027Tax refunded or set off\u0027 further confirm this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax years, and HMRC-related processes, which are key indicators of corporation tax documents. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax calculation summary,\u0027 and \u0027Tax refunded or set off\u0027 further confirm 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": 0.1392857142857143,
"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": 2555
},
{
"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 contains references to tax-related information, such as \u0027Tax Reference\u0027, \u0027Your tax adviser\u0027, and \u0027SA 100\u0027, which is a self-assessment tax return form in the UK. 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-DZFmS8aSblHRVflvtjKoyEsi3uufe\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 contains references to tax-related information, such as \\\u0027Tax Reference\\\u0027, \\\u0027Your tax adviser\\\u0027, and \\\u0027SA 100\\\u0027, which is a self-assessment tax return form in the UK. 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=1777294512, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=213, prompt_tokens=1179, total_tokens=1392, 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_"
},
"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, \u0027Your tax adviser\u0027, and \u0027SA 100\u0027, which is a self-assessment tax return form in the UK. These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to tax-related information, such as \u0027Tax Reference\u0027, \u0027Your tax adviser\u0027, and \u0027SA 100\u0027, which is a self-assessment tax return form in the UK. 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": 0.065,
"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": 577
},
{
"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 contains references to a tax return (SA100), tax reference numbers, and declarations related to tax filing. 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-DZFmUUqo3TpiDM3FKiekodR68scHv\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 contains references to a tax return (SA100), tax reference numbers, and declarations related to tax filing. 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=1777294514, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=197, prompt_tokens=1264, total_tokens=1461, 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_tokens=0)), prompt_filter_results=[{\u0027prompt_index\u0027: 0, \u0027conten"
},
"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 numbers, and declarations related to tax filing. These are key indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to a tax return (SA100), tax reference numbers, and declarations related to tax filing. 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": "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": 1071
},
{
"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, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment tax reporting, which aligns with corporation tax filings or related documentation.\",\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-DZFmVam18MiBY23ek3o7pahMUGaCj\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, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment tax reporting, which aligns with corporation tax filings or related documentation.\",\\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=1777294515, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=217, prompt_tokens=1566, total_tokens=1783, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details"
},
"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, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment tax reporting, which aligns with corporation tax filings or related documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment tax reporting, which aligns with corporation tax filings or related documentation.\",\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": 1833
}
],
"start_page": 1
},
{
"avg_confidence": 0.9,
"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": 0.9,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data such as business expenses, disallowable expenses, and total expenses, which are typical components of a financial statement. It does not match the structure or content of other categories like invoices, payroll, or loan statements.\",\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-DZFmXXYK5wN7EbLbGu56LNxipnxI4\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains detailed financial data such as business expenses, disallowable expenses, and total expenses, which are typical components of a financial statement. It does not match the structure or content of other categories like invoices, payroll, or loan statements.\",\\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=1777294517, 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=1586, total_tokens=1798, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=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 detailed financial data such as business expenses, disallowable expenses, and total expenses, which are typical components of a financial statement. It does not match the structure or content of other categories like invoices, payroll, or loan statements.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data such as business expenses, disallowable expenses, and total expenses, which are typical components of a financial statement. It does not match the structure or content of other categories like invoices, payroll, or loan statements.\",\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": 1495
}
],
"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 calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., \u0027box 47\u0027, \u0027box 63\u0027). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return or related document. 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-DZFmYzDe70MiXCobsZtLwFtBmxcu2\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 calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., \\\u0027box 47\\\u0027, \\\u0027box 63\\\u0027). Additionally, it includes a tax reference number and mentions \\\u0027Self-employment (full) notes,\\\u0027 which are indicative of a tax return or related document. 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=1777294518, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=247, prompt_tokens=1753, total_tokens=2000, 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 references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., \u0027box 47\u0027, \u0027box 63\u0027). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return or related document. 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 calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., \u0027box 47\u0027, \u0027box 63\u0027). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return or related document. 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": 2581
},
{
"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 profits, adjustments for accounting periods, and loss carry-forwards, which are key indicators of a corporation tax document. Specific terms like \u0027taxable profit or loss,\u0027 \u0027Self-employment (full) notes,\u0027 and \u0027adjusted profit for 2024-25\u0027 strongly suggest this is related to corporation tax filings, likely part of a self-assessment or CT600 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-DZFmauqlo0sH9Ckm2EdtCgtUlOOnq\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 profits, adjustments for accounting periods, and loss carry-forwards, which are key indicators of a corporation tax document. Specific terms like \\\u0027taxable profit or loss,\\\u0027 \\\u0027Self-employment (full) notes,\\\u0027 and \\\u0027adjusted profit for 2024-25\\\u0027 strongly suggest this is related to corporation tax filings, likely part of a self-assessment or CT600 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=1777294520, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=245, prompt_tokens=1735, total_tokens=1980, completion_tokens_details=CompletionTokensDetails(accep"
},
"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 profits, adjustments for accounting periods, and loss carry-forwards, which are key indicators of a corporation tax document. Specific terms like \u0027taxable profit or loss,\u0027 \u0027Self-employment (full) notes,\u0027 and \u0027adjusted profit for 2024-25\u0027 strongly suggest this is related to corporation tax filings, likely part of a self-assessment or CT600 form.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profits, adjustments for accounting periods, and loss carry-forwards, which are key indicators of a corporation tax document. Specific terms like \u0027taxable profit or loss,\u0027 \u0027Self-employment (full) notes,\u0027 and \u0027adjusted profit for 2024-25\u0027 strongly suggest this is related to corporation tax filings, likely part of a self-assessment or CT600 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": "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": 2614
}
],
"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, which are key indicators of a financial statement. Terms like \u0027Net business assets,\u0027 \u0027Balance at start of period,\u0027 and \u0027Capital account\u0027 further confirm 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-DZFmcmyXJOoUQcRngWlcSFP05tr8n\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, which are key indicators of a financial statement. Terms like \\\u0027Net business assets,\\\u0027 \\\u0027Balance at start of period,\\\u0027 and \\\u0027Capital account\\\u0027 further confirm 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=1777294522, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=213, prompt_tokens=1545, total_tokens=1758, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_de"
},
"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, which are key indicators of a financial statement. Terms like \u0027Net business assets,\u0027 \u0027Balance at start of period,\u0027 and \u0027Capital account\u0027 further confirm this classification.",
"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, which are key indicators of a financial statement. Terms like \u0027Net business assets,\u0027 \u0027Balance at start of period,\u0027 and \u0027Capital account\u0027 further confirm 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": "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": 1436
}
],
"start_page": 15
},
{
"avg_confidence": 0.9375,
"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": 19,
"pages": [
{
"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-related information, including Class 2 and Class 4 National Insurance contributions, tax reference numbers, and self-employment profits. These are key indicators of a corporation tax-related 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-DZFmdnfGHEN6BtMdy914WQsnDrqO8\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-related information, including Class 2 and Class 4 National Insurance contributions, tax reference numbers, and self-employment profits. These are key indicators of a corporation tax-related 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=1777294523, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=206, prompt_tokens=1211, total_tokens=1417, 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"
},
"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 Class 2 and Class 4 National Insurance contributions, tax reference numbers, and self-employment profits. These are key indicators of a corporation tax-related document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax-related information, including Class 2 and Class 4 National Insurance contributions, tax reference numbers, and self-employment profits. These are key indicators of a corporation tax-related 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": 0.06666666666666667,
"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": 766
},
{
"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, tax year, Unique Taxpayer Reference (UTR), and mentions of payments on account, underpaid tax, and other tax-related details. 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-DZFmeGLvQTCGlheulWuBcfXjrpzKq\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, tax year, Unique Taxpayer Reference (UTR), and mentions of payments on account, underpaid tax, and other tax-related details. 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=1777294524, 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=1682, total_tokens=1896, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(audio_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 HMRC, tax calculation summary, tax year, Unique Taxpayer Reference (UTR), and mentions of payments on account, underpaid tax, and other tax-related details. 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 HMRC, tax calculation summary, tax year, Unique Taxpayer Reference (UTR), and mentions of payments on account, underpaid tax, and other tax-related details. 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": 0.17505882352941174,
"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": 17,
"text_length": 2275
},
{
"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, \u0027Blind person\u0027s surplus allowance\u0027, \u0027married couple\u0027s surplus allowance\u0027, \u0027Adjustments to tax due\u0027, and \u0027self-assessment statement of account\u0027. These are key indicators of a corporation tax document, specifically related to self-assessment and tax adjustments.\",\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-DZFmgOLV9zd0ABtVakbH2LeHFoFd0\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, \\\u0027Blind person\\\u0027s surplus allowance\\\u0027, \\\u0027married couple\\\u0027s surplus allowance\\\u0027, \\\u0027Adjustments to tax due\\\u0027, and \\\u0027self-assessment statement of account\\\u0027. These are key indicators of a corporation tax document, specifically related to self-assessment and tax adjustments.\",\\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=1777294526, 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=1329, total_tokens=1557, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, rea"
},
"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, \u0027Blind person\u0027s surplus allowance\u0027, \u0027married couple\u0027s surplus allowance\u0027, \u0027Adjustments to tax due\u0027, and \u0027self-assessment statement of account\u0027. These are key indicators of a corporation tax document, specifically related to self-assessment and tax adjustments.",
"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, \u0027Blind person\u0027s surplus allowance\u0027, \u0027married couple\u0027s surplus allowance\u0027, \u0027Adjustments to tax due\u0027, and \u0027self-assessment statement of account\u0027. These are key indicators of a corporation tax document, specifically related to self-assessment and tax adjustments.\",\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": 18,
"text_length": 1362
},
{
"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 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-DZFmh3DABjNWpeExAKgxVD9UjDPkW\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 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=1777294527, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=217, prompt_tokens=1280, total_tokens=1497, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_deta"
},
"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 corporation tax document.",
"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 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": 0.11666666666666667,
"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": 0.08333333333333333,
"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": 19,
"text_length": 878
}
],
"start_page": 16
},
{
"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": 21,
"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-DZFmiijxYHzXVyzR5TpWHd8kNTRYm\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=1777294528, 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=1961, total_tokens=2175, 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": 20,
"text_length": 1708
},
{
"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 references to capital gains (e.g., CGTg, CGTb, CGTi, CGTr, CGTc), which are typically part of financial statements or tax-related summaries. It does not include transaction details, loan references, or invoice-specific identifiers, which rules out 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-DZFmkutT9L9MZlYbFLb7Tu7r0Uu2q\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 references to capital gains (e.g., CGTg, CGTb, CGTi, CGTr, CGTc), which are typically part of financial statements or tax-related summaries. It does not include transaction details, loan references, or invoice-specific identifiers, which rules out 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=1777294530, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=222, prompt_tokens=1117, total_tokens=1339, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_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 capital gains (e.g., CGTg, CGTb, CGTi, CGTr, CGTc), which are typically part of financial statements or tax-related summaries. It does not include transaction details, loan references, or invoice-specific identifiers, which rules out other categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to capital gains (e.g., CGTg, CGTb, CGTi, CGTr, CGTc), which are typically part of financial statements or tax-related summaries. It does not include transaction details, loan references, or invoice-specific identifiers, which rules out 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": 0.03333333333333333,
"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": 21,
"text_length": 402
}
],
"start_page": 20
},
{
"avg_confidence": 0.9,
"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": 22,
"pages": [
{
"confidence_score": 0.9,
"document_type": "p32_payroll",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.9,\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. The presence of specific payment dates (31/01/26 and 31/07/26) and amounts due (e.g., \u00a332.76) 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-DZFmlSv4jwfDStGhYW5b7uoFOufPj\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"p32_payroll\",\\n \"confidence\": 0.9,\\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. The presence of specific payment dates (31/01/26 and 31/07/26) and amounts due (e.g., \u00a332.76) 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=1777294531, 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=1137, total_tokens=1379, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_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 \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. The presence of specific payment dates (31/01/26 and 31/07/26) and amounts due (e.g., \u00a332.76) further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.9,\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. The presence of specific payment dates (31/01/26 and 31/07/26) and amounts due (e.g., \u00a332.76) 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": "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": 22,
"text_length": 400
}
],
"start_page": 22
},
{
"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": "corporation_tax",
"end_page": 24,
"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 mentions \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of tax-related calculations. Additionally, the reference to a specific tax year (\u0027Year ended 5 April 2025\u0027) aligns with corporation tax or personal tax documents. However, the absence of explicit references to CT600 forms or corporation-specific terms slightly reduces confidence.\",\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-DZFmn0IplnEyd2CX9D0hdgIIcpUN1\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 mentions \\\u0027Tax Due After Notional Taxes Paid,\\\u0027 \\\u0027Class 4 NIC Due,\\\u0027 and \\\u0027Payments on Account,\\\u0027 which are indicative of tax-related calculations. Additionally, the reference to a specific tax year (\\\u0027Year ended 5 April 2025\\\u0027) aligns with corporation tax or personal tax documents. However, the absence of explicit references to CT600 forms or corporation-specific terms slightly reduces confidence.\",\\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=1777294533, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=243, prompt_tokens=1111, total_tokens=1354, completion_tokens_details=CompletionT"
},
"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 \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of tax-related calculations. Additionally, the reference to a specific tax year (\u0027Year ended 5 April 2025\u0027) aligns with corporation tax or personal tax documents. However, the absence of explicit references to CT600 forms or corporation-specific terms slightly reduces confidence.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document mentions \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of tax-related calculations. Additionally, the reference to a specific tax year (\u0027Year ended 5 April 2025\u0027) aligns with corporation tax or personal tax documents. However, the absence of explicit references to CT600 forms or corporation-specific terms slightly reduces confidence.\",\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": 23,
"text_length": 374
},
{
"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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of specific tax terms and calculations aligns 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-DZFmoIwWphUpj6XW1sFQ1pfRzrLMB\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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of specific tax terms and calculations aligns 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=1777294534, 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=1119, total_tokens=1333, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_de"
},
"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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of specific tax terms and calculations aligns with corporation tax or self-assessment tax documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document references Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of specific tax terms and calculations aligns 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": 24,
"text_length": 359
}
],
"start_page": 23
},
{
"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": 27,
"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, balances, or account numbers typical of bank or loan statements, nor are there invoice numbers, property details, or payroll 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-DZFmrMa7sUhG0JligRIErBZ5Uqpxe\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, balances, or account numbers typical of bank or loan statements, nor are there invoice numbers, property details, or payroll 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=1777294537, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=233, prompt_tokens=1410, total_tokens=1643, completion_tokens_details=CompletionTokensDetails(accept"
},
"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, balances, or account numbers typical of bank or loan statements, nor are there invoice numbers, property details, or payroll 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, balances, or account numbers typical of bank or loan statements, nor are there invoice numbers, property details, or payroll 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": "both_agree",
"pattern_based": {
"confidence": 0.1616842105263158,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.0,
"total_score": 0.12705882352941175
},
"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": "financial_statement"
}
},
"page_number": 26,
"text_length": 1304
},
{
"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 net business assets. It also includes a capital account summary and adjustments to arrive at profit/loss for tax purposes, which are key indicators of a financial 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 }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFmsfjxpXmjxzESoSAmldod13NnR\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 net business assets. It also includes a capital account summary and adjustments to arrive at profit/loss for tax purposes, which are key indicators of a financial 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 }\\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=1777294538, 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=1378, total_tokens=1590, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=Promp"
},
"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 net business assets. It also includes a capital account summary and adjustments to arrive at profit/loss for tax purposes, which are key indicators of a financial statement.",
"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 net business assets. It also includes a capital account summary and adjustments to arrive at profit/loss for tax purposes, which are key indicators of a financial 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 }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.25768421052631574,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"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": 0.15789473684210525,
"pattern_score": 0.3,
"total_score": 0.25768421052631574
},
"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": 0.0,
"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": 27,
"text_length": 1111
}
],
"start_page": 26
}
],
"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 references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, it includes details about tax return submission and approval, which aligns with corporation tax documentation.\",\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-DZFmEoZMR7aG2qef5lUzPGRLoZ4sc\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 references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, it includes details about tax return submission and approval, which aligns with corporation tax documentation.\",\\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=1777294498, 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=1385, total_tokens=1609, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_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 references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, it includes details about tax return submission and approval, which aligns with corporation tax documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, it includes details about tax return submission and approval, which aligns with corporation tax documentation.\",\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": 1900
},
{
"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 mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are indicative of a corporation tax document or a tax return filing.\",\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-DZFmGJXel3Jkho0CfQU5iqym9hrxm\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 mentions \\\u00272024/25 Tax Return\\\u0027 and includes references to tax-related schedules such as \\\u0027Rate Band Analysis\\\u0027, \\\u0027Payments on Account\\\u0027, and \\\u0027Class 4 National Insurance Charge\\\u0027. These are indicative of a corporation tax document or a tax return filing.\",\\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=1777294500, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=216, prompt_tokens=1040, total_tokens=1256, 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 mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are indicative of a corporation tax document or a tax return filing.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are indicative of a corporation tax document or a tax return filing.\",\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": 2,
"text_length": 213
},
{
"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, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. It also includes details about tax deadlines, penalties, and instructions for filing a tax return, which are specific to corporation tax filings.\",\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-DZFmH0uKZsNJQ7Gk3df6URWKdNdZe\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, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. It also includes details about tax deadlines, penalties, and instructions for filing a tax return, which are specific to corporation tax filings.\",\\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=1777294501, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=223, prompt_tokens=1737, total_tokens=1960, 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 HMRC, tax return, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. It also includes details about tax deadlines, penalties, and instructions for filing a tax return, which are specific to corporation tax filings.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax return, UTR (Unique Taxpayer Reference), and Self Assessment, which are key indicators of a corporation tax document. It also includes details about tax deadlines, penalties, and instructions for filing a tax return, which are specific to corporation tax filings.\",\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": 2711
},
{
"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 returns, taxable income, supplementary pages for employment, self-employment, capital gains, and UK property income. These are key indicators of a corporation tax-related document, specifically a tax return or related 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-DZFmJmlf4eDtBQwzgv8ldO2zU5kcD\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 returns, taxable income, supplementary pages for employment, self-employment, capital gains, and UK property income. These are key indicators of a corporation tax-related document, specifically a tax return or related 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=1777294503, 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=1731, total_tokens=1941, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails("
},
"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, taxable income, supplementary pages for employment, self-employment, capital gains, and UK property income. These are key indicators of a corporation tax-related document, specifically a tax return or related form.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax returns, taxable income, supplementary pages for employment, self-employment, capital gains, and UK property income. These are key indicators of a corporation tax-related document, specifically a tax return or related 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": 0.9,
"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-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Dividends\u0027, \u0027State Pension\u0027, and \u0027Taxable Incapacity Benefit\u0027. These are indicative of a tax return or corporation tax document. Additionally, the presence of \u0027SA100\u0027 suggests it is related to a self-assessment tax return form, which aligns with corporation tax filings.\",\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-DZFmLOxdUYYph76n63QsiqqRFRYPI\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-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Taxed UK interest\\\u0027, \\\u0027Untaxed UK interest\\\u0027, \\\u0027Dividends\\\u0027, \\\u0027State Pension\\\u0027, and \\\u0027Taxable Incapacity Benefit\\\u0027. These are indicative of a tax return or corporation tax document. Additionally, the presence of \\\u0027SA100\\\u0027 suggests it is related to a self-assessment tax return form, which aligns with corporation tax filings.\",\\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=1777294505, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=249, prompt_tokens=1631, total_tokens=1880, completion_tokens_details=CompletionT"
},
"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, \u0027Dividends\u0027, \u0027State Pension\u0027, and \u0027Taxable Incapacity Benefit\u0027. These are indicative of a tax return or corporation tax document. Additionally, the presence of \u0027SA100\u0027 suggests it is related to a self-assessment tax return form, which aligns with corporation tax filings.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Dividends\u0027, \u0027State Pension\u0027, and \u0027Taxable Incapacity Benefit\u0027. These are indicative of a tax return or corporation tax document. Additionally, the presence of \u0027SA100\u0027 suggests it is related to a self-assessment tax return form, which aligns with corporation tax filings.\",\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": 2225
},
{
"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 allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\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-DZFmMpN905G7ireyJXvSJPQCMLMAl\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 allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \\\u0027Tax Reference\\\u0027, \\\u0027Gift Aid\\\u0027, and \\\u0027SA 100\\\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\\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=1777294506, 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=1597, total_tokens=1824, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning"
},
"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 allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.",
"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 allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\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": 2173
},
{
"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 contains references to tax-related information, such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, and \u0027High Income Child Benefit Charge\u0027. These are indicative of a tax-related document, specifically related to personal or corporate tax filings. The presence of terms like \u0027SA 100\u0027 and detailed tax-related sections 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-DZFmPyMhOwLR3unO5JIEmJT7MDL0p\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 contains references to tax-related information, such as \\\u0027Tax Reference\\\u0027, \\\u0027Student Loan and Postgraduate Loan repayments\\\u0027, and \\\u0027High Income Child Benefit Charge\\\u0027. These are indicative of a tax-related document, specifically related to personal or corporate tax filings. The presence of terms like \\\u0027SA 100\\\u0027 and detailed tax-related sections 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=1777294509, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=233, prompt_tokens=1537, total_tokens=1770, completion_tokens_details=CompletionTokensDetails(acce"
},
"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, and \u0027High Income Child Benefit Charge\u0027. These are indicative of a tax-related document, specifically related to personal or corporate tax filings. The presence of terms like \u0027SA 100\u0027 and detailed tax-related sections further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to tax-related information, such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, and \u0027High Income Child Benefit Charge\u0027. These are indicative of a tax-related document, specifically related to personal or corporate tax filings. The presence of terms like \u0027SA 100\u0027 and detailed tax-related sections 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": 0.11454545454545456,
"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": 2182
},
{
"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 returns, tax calculations, tax years, and HMRC-related processes, which are key indicators of corporation tax documents. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax calculation summary,\u0027 and \u0027Tax refunded or set off\u0027 further confirm 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-DZFmRIpadLO4DCmlHymiJXyoplcjE\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 returns, tax calculations, tax years, and HMRC-related processes, which are key indicators of corporation tax documents. Specific mentions of \\\u0027Self Assessment Statement,\\\u0027 \\\u0027Tax calculation summary,\\\u0027 and \\\u0027Tax refunded or set off\\\u0027 further confirm 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=1777294511, 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=1676, total_tokens=1895, 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 returns, tax calculations, tax years, and HMRC-related processes, which are key indicators of corporation tax documents. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax calculation summary,\u0027 and \u0027Tax refunded or set off\u0027 further confirm this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax years, and HMRC-related processes, which are key indicators of corporation tax documents. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax calculation summary,\u0027 and \u0027Tax refunded or set off\u0027 further confirm 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": 0.1392857142857143,
"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": 2555
},
{
"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 contains references to tax-related information, such as \u0027Tax Reference\u0027, \u0027Your tax adviser\u0027, and \u0027SA 100\u0027, which is a self-assessment tax return form in the UK. 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-DZFmS8aSblHRVflvtjKoyEsi3uufe\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 contains references to tax-related information, such as \\\u0027Tax Reference\\\u0027, \\\u0027Your tax adviser\\\u0027, and \\\u0027SA 100\\\u0027, which is a self-assessment tax return form in the UK. 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=1777294512, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=213, prompt_tokens=1179, total_tokens=1392, 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_"
},
"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, \u0027Your tax adviser\u0027, and \u0027SA 100\u0027, which is a self-assessment tax return form in the UK. These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to tax-related information, such as \u0027Tax Reference\u0027, \u0027Your tax adviser\u0027, and \u0027SA 100\u0027, which is a self-assessment tax return form in the UK. 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": 0.065,
"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": 577
},
{
"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 contains references to a tax return (SA100), tax reference numbers, and declarations related to tax filing. 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-DZFmUUqo3TpiDM3FKiekodR68scHv\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 contains references to a tax return (SA100), tax reference numbers, and declarations related to tax filing. 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=1777294514, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=197, prompt_tokens=1264, total_tokens=1461, 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_tokens=0)), prompt_filter_results=[{\u0027prompt_index\u0027: 0, \u0027conten"
},
"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 numbers, and declarations related to tax filing. These are key indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to a tax return (SA100), tax reference numbers, and declarations related to tax filing. 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": "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": 1071
},
{
"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, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment tax reporting, which aligns with corporation tax filings or related documentation.\",\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-DZFmVam18MiBY23ek3o7pahMUGaCj\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, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment tax reporting, which aligns with corporation tax filings or related documentation.\",\\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=1777294515, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=217, prompt_tokens=1566, total_tokens=1783, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details"
},
"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, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment tax reporting, which aligns with corporation tax filings or related documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment tax reporting, which aligns with corporation tax filings or related documentation.\",\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": 1833
},
{
"confidence_score": 0.9,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data such as business expenses, disallowable expenses, and total expenses, which are typical components of a financial statement. It does not match the structure or content of other categories like invoices, payroll, or loan statements.\",\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-DZFmXXYK5wN7EbLbGu56LNxipnxI4\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains detailed financial data such as business expenses, disallowable expenses, and total expenses, which are typical components of a financial statement. It does not match the structure or content of other categories like invoices, payroll, or loan statements.\",\\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=1777294517, 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=1586, total_tokens=1798, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=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 detailed financial data such as business expenses, disallowable expenses, and total expenses, which are typical components of a financial statement. It does not match the structure or content of other categories like invoices, payroll, or loan statements.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data such as business expenses, disallowable expenses, and total expenses, which are typical components of a financial statement. It does not match the structure or content of other categories like invoices, payroll, or loan statements.\",\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": 1495
},
{
"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 calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., \u0027box 47\u0027, \u0027box 63\u0027). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return or related document. 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-DZFmYzDe70MiXCobsZtLwFtBmxcu2\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 calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., \\\u0027box 47\\\u0027, \\\u0027box 63\\\u0027). Additionally, it includes a tax reference number and mentions \\\u0027Self-employment (full) notes,\\\u0027 which are indicative of a tax return or related document. 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=1777294518, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=247, prompt_tokens=1753, total_tokens=2000, 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 references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., \u0027box 47\u0027, \u0027box 63\u0027). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return or related document. 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 calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., \u0027box 47\u0027, \u0027box 63\u0027). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return or related document. 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": 2581
},
{
"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 profits, adjustments for accounting periods, and loss carry-forwards, which are key indicators of a corporation tax document. Specific terms like \u0027taxable profit or loss,\u0027 \u0027Self-employment (full) notes,\u0027 and \u0027adjusted profit for 2024-25\u0027 strongly suggest this is related to corporation tax filings, likely part of a self-assessment or CT600 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-DZFmauqlo0sH9Ckm2EdtCgtUlOOnq\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 profits, adjustments for accounting periods, and loss carry-forwards, which are key indicators of a corporation tax document. Specific terms like \\\u0027taxable profit or loss,\\\u0027 \\\u0027Self-employment (full) notes,\\\u0027 and \\\u0027adjusted profit for 2024-25\\\u0027 strongly suggest this is related to corporation tax filings, likely part of a self-assessment or CT600 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=1777294520, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=245, prompt_tokens=1735, total_tokens=1980, completion_tokens_details=CompletionTokensDetails(accep"
},
"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 profits, adjustments for accounting periods, and loss carry-forwards, which are key indicators of a corporation tax document. Specific terms like \u0027taxable profit or loss,\u0027 \u0027Self-employment (full) notes,\u0027 and \u0027adjusted profit for 2024-25\u0027 strongly suggest this is related to corporation tax filings, likely part of a self-assessment or CT600 form.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profits, adjustments for accounting periods, and loss carry-forwards, which are key indicators of a corporation tax document. Specific terms like \u0027taxable profit or loss,\u0027 \u0027Self-employment (full) notes,\u0027 and \u0027adjusted profit for 2024-25\u0027 strongly suggest this is related to corporation tax filings, likely part of a self-assessment or CT600 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": "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": 2614
},
{
"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, which are key indicators of a financial statement. Terms like \u0027Net business assets,\u0027 \u0027Balance at start of period,\u0027 and \u0027Capital account\u0027 further confirm 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-DZFmcmyXJOoUQcRngWlcSFP05tr8n\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, which are key indicators of a financial statement. Terms like \\\u0027Net business assets,\\\u0027 \\\u0027Balance at start of period,\\\u0027 and \\\u0027Capital account\\\u0027 further confirm 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=1777294522, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=213, prompt_tokens=1545, total_tokens=1758, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_de"
},
"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, which are key indicators of a financial statement. Terms like \u0027Net business assets,\u0027 \u0027Balance at start of period,\u0027 and \u0027Capital account\u0027 further confirm this classification.",
"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, which are key indicators of a financial statement. Terms like \u0027Net business assets,\u0027 \u0027Balance at start of period,\u0027 and \u0027Capital account\u0027 further confirm 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": "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": 1436
},
{
"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-related information, including Class 2 and Class 4 National Insurance contributions, tax reference numbers, and self-employment profits. These are key indicators of a corporation tax-related 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-DZFmdnfGHEN6BtMdy914WQsnDrqO8\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-related information, including Class 2 and Class 4 National Insurance contributions, tax reference numbers, and self-employment profits. These are key indicators of a corporation tax-related 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=1777294523, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=206, prompt_tokens=1211, total_tokens=1417, 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"
},
"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 Class 2 and Class 4 National Insurance contributions, tax reference numbers, and self-employment profits. These are key indicators of a corporation tax-related document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax-related information, including Class 2 and Class 4 National Insurance contributions, tax reference numbers, and self-employment profits. These are key indicators of a corporation tax-related 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": 0.06666666666666667,
"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": 766
},
{
"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, tax year, Unique Taxpayer Reference (UTR), and mentions of payments on account, underpaid tax, and other tax-related details. 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-DZFmeGLvQTCGlheulWuBcfXjrpzKq\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, tax year, Unique Taxpayer Reference (UTR), and mentions of payments on account, underpaid tax, and other tax-related details. 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=1777294524, 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=1682, total_tokens=1896, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(audio_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 HMRC, tax calculation summary, tax year, Unique Taxpayer Reference (UTR), and mentions of payments on account, underpaid tax, and other tax-related details. 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 HMRC, tax calculation summary, tax year, Unique Taxpayer Reference (UTR), and mentions of payments on account, underpaid tax, and other tax-related details. 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": 0.17505882352941174,
"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": 17,
"text_length": 2275
},
{
"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, \u0027Blind person\u0027s surplus allowance\u0027, \u0027married couple\u0027s surplus allowance\u0027, \u0027Adjustments to tax due\u0027, and \u0027self-assessment statement of account\u0027. These are key indicators of a corporation tax document, specifically related to self-assessment and tax adjustments.\",\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-DZFmgOLV9zd0ABtVakbH2LeHFoFd0\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, \\\u0027Blind person\\\u0027s surplus allowance\\\u0027, \\\u0027married couple\\\u0027s surplus allowance\\\u0027, \\\u0027Adjustments to tax due\\\u0027, and \\\u0027self-assessment statement of account\\\u0027. These are key indicators of a corporation tax document, specifically related to self-assessment and tax adjustments.\",\\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=1777294526, 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=1329, total_tokens=1557, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, rea"
},
"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, \u0027Blind person\u0027s surplus allowance\u0027, \u0027married couple\u0027s surplus allowance\u0027, \u0027Adjustments to tax due\u0027, and \u0027self-assessment statement of account\u0027. These are key indicators of a corporation tax document, specifically related to self-assessment and tax adjustments.",
"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, \u0027Blind person\u0027s surplus allowance\u0027, \u0027married couple\u0027s surplus allowance\u0027, \u0027Adjustments to tax due\u0027, and \u0027self-assessment statement of account\u0027. These are key indicators of a corporation tax document, specifically related to self-assessment and tax adjustments.\",\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": 18,
"text_length": 1362
},
{
"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 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-DZFmh3DABjNWpeExAKgxVD9UjDPkW\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 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=1777294527, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=217, prompt_tokens=1280, total_tokens=1497, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_deta"
},
"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 corporation tax document.",
"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 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": 0.11666666666666667,
"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": 0.08333333333333333,
"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": 19,
"text_length": 878
},
{
"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-DZFmiijxYHzXVyzR5TpWHd8kNTRYm\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=1777294528, 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=1961, total_tokens=2175, 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": 20,
"text_length": 1708
},
{
"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 references to capital gains (e.g., CGTg, CGTb, CGTi, CGTr, CGTc), which are typically part of financial statements or tax-related summaries. It does not include transaction details, loan references, or invoice-specific identifiers, which rules out 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-DZFmkutT9L9MZlYbFLb7Tu7r0Uu2q\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 references to capital gains (e.g., CGTg, CGTb, CGTi, CGTr, CGTc), which are typically part of financial statements or tax-related summaries. It does not include transaction details, loan references, or invoice-specific identifiers, which rules out 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=1777294530, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=222, prompt_tokens=1117, total_tokens=1339, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_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 capital gains (e.g., CGTg, CGTb, CGTi, CGTr, CGTc), which are typically part of financial statements or tax-related summaries. It does not include transaction details, loan references, or invoice-specific identifiers, which rules out other categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to capital gains (e.g., CGTg, CGTb, CGTi, CGTr, CGTc), which are typically part of financial statements or tax-related summaries. It does not include transaction details, loan references, or invoice-specific identifiers, which rules out 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": 0.03333333333333333,
"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": 21,
"text_length": 402
},
{
"confidence_score": 0.9,
"document_type": "p32_payroll",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.9,\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. The presence of specific payment dates (31/01/26 and 31/07/26) and amounts due (e.g., \u00a332.76) 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-DZFmlSv4jwfDStGhYW5b7uoFOufPj\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"p32_payroll\",\\n \"confidence\": 0.9,\\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. The presence of specific payment dates (31/01/26 and 31/07/26) and amounts due (e.g., \u00a332.76) 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=1777294531, 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=1137, total_tokens=1379, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_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 \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. The presence of specific payment dates (31/01/26 and 31/07/26) and amounts due (e.g., \u00a332.76) further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.9,\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. The presence of specific payment dates (31/01/26 and 31/07/26) and amounts due (e.g., \u00a332.76) 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": "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": 22,
"text_length": 400
},
{
"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 mentions \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of tax-related calculations. Additionally, the reference to a specific tax year (\u0027Year ended 5 April 2025\u0027) aligns with corporation tax or personal tax documents. However, the absence of explicit references to CT600 forms or corporation-specific terms slightly reduces confidence.\",\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-DZFmn0IplnEyd2CX9D0hdgIIcpUN1\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 mentions \\\u0027Tax Due After Notional Taxes Paid,\\\u0027 \\\u0027Class 4 NIC Due,\\\u0027 and \\\u0027Payments on Account,\\\u0027 which are indicative of tax-related calculations. Additionally, the reference to a specific tax year (\\\u0027Year ended 5 April 2025\\\u0027) aligns with corporation tax or personal tax documents. However, the absence of explicit references to CT600 forms or corporation-specific terms slightly reduces confidence.\",\\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=1777294533, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=243, prompt_tokens=1111, total_tokens=1354, completion_tokens_details=CompletionT"
},
"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 \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of tax-related calculations. Additionally, the reference to a specific tax year (\u0027Year ended 5 April 2025\u0027) aligns with corporation tax or personal tax documents. However, the absence of explicit references to CT600 forms or corporation-specific terms slightly reduces confidence.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document mentions \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account,\u0027 which are indicative of tax-related calculations. Additionally, the reference to a specific tax year (\u0027Year ended 5 April 2025\u0027) aligns with corporation tax or personal tax documents. However, the absence of explicit references to CT600 forms or corporation-specific terms slightly reduces confidence.\",\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": 23,
"text_length": 374
},
{
"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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of specific tax terms and calculations aligns 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-DZFmoIwWphUpj6XW1sFQ1pfRzrLMB\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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of specific tax terms and calculations aligns 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=1777294534, 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=1119, total_tokens=1333, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_de"
},
"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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of specific tax terms and calculations aligns with corporation tax or self-assessment tax documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document references Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of specific tax terms and calculations aligns 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": 24,
"text_length": 359
},
{
"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 personal allowances and a year-end date, which are typical components of financial summaries or statements. However, it lacks specific indicators for other categories such as invoices, payroll, or loan-related documents. The absence of transaction details, 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-DZFmqiPeOyjTqjWWUt5Qhb8SUeI2x\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 personal allowances and a year-end date, which are typical components of financial summaries or statements. However, it lacks specific indicators for other categories such as invoices, payroll, or loan-related documents. The absence of transaction details, 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=1777294536, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=223, prompt_tokens=1057, total_tokens=1280, completion_tokens_details=CompletionTokensDetails(accepted_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 references to personal allowances and a year-end date, which are typical components of financial summaries or statements. However, it lacks specific indicators for other categories such as invoices, payroll, or loan-related documents. The absence of transaction details, 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 contains references to personal allowances and a year-end date, which are typical components of financial summaries or statements. However, it lacks specific indicators for other categories such as invoices, payroll, or loan-related documents. The absence of transaction details, 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": 25,
"text_length": 184
},
{
"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, balances, or account numbers typical of bank or loan statements, nor are there invoice numbers, property details, or payroll 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-DZFmrMa7sUhG0JligRIErBZ5Uqpxe\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, balances, or account numbers typical of bank or loan statements, nor are there invoice numbers, property details, or payroll 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=1777294537, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=233, prompt_tokens=1410, total_tokens=1643, completion_tokens_details=CompletionTokensDetails(accept"
},
"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, balances, or account numbers typical of bank or loan statements, nor are there invoice numbers, property details, or payroll 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, balances, or account numbers typical of bank or loan statements, nor are there invoice numbers, property details, or payroll 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": "both_agree",
"pattern_based": {
"confidence": 0.1616842105263158,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.0,
"total_score": 0.12705882352941175
},
"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": "financial_statement"
}
},
"page_number": 26,
"text_length": 1304
},
{
"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 net business assets. It also includes a capital account summary and adjustments to arrive at profit/loss for tax purposes, which are key indicators of a financial 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 }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFmsfjxpXmjxzESoSAmldod13NnR\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 net business assets. It also includes a capital account summary and adjustments to arrive at profit/loss for tax purposes, which are key indicators of a financial 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 }\\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=1777294538, 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=1378, total_tokens=1590, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=Promp"
},
"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 net business assets. It also includes a capital account summary and adjustments to arrive at profit/loss for tax purposes, which are key indicators of a financial statement.",
"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 net business assets. It also includes a capital account summary and adjustments to arrive at profit/loss for tax purposes, which are key indicators of a financial 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 }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.25768421052631574,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"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": 0.15789473684210525,
"pattern_score": 0.3,
"total_score": 0.25768421052631574
},
"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": 0.0,
"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": 27,
"text_length": 1111
},
{
"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 information about residence status, domicile, and dual residence status for a specific individual and tax year. These details are typically part of a financial or tax-related summary, which aligns most closely with a financial statement. However, the lack of explicit financial figures or balance sheet details slightly reduces confidence.\",\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-DZFmuJUhweQs2V5Xo1lPrayXu1RPn\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 information about residence status, domicile, and dual residence status for a specific individual and tax year. These details are typically part of a financial or tax-related summary, which aligns most closely with a financial statement. However, the lack of explicit financial figures or balance sheet details slightly reduces confidence.\",\\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=1777294540, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=222, prompt_tokens=1087, total_tokens=1309, 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 information about residence status, domicile, and dual residence status for a specific individual and tax year. These details are typically part of a financial or tax-related summary, which aligns most closely with a financial statement. However, the lack of explicit financial figures or balance sheet details slightly reduces confidence.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.8,\n \"reasoning\": \"The document contains information about residence status, domicile, and dual residence status for a specific individual and tax year. These details are typically part of a financial or tax-related summary, which aligns most closely with a financial statement. However, the lack of explicit financial figures or balance sheet details slightly reduces confidence.\",\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": 28,
"text_length": 327
}
],
"processing_timestamp": "2026-04-27T12:55:41.763947"
}