| Original Filename: | 1THO007_Thomas_M_SA100_24-25_1778503609.pdf |
| Upload Date: | May 11, 2026 at 12:56 PM |
| Total Pages: | 28 |
| Processing Status: | Completed |
| Total Documents: | 14 |
| Corporation Taxs: | 8 |
| Financial Statements: | 5 |
| P32 Payrolls: | 1 |
Corporation Tax
95%Corporation Tax
90%Corporation Tax
95%Corporation Tax
95%Corporation Tax
95%Corporation Tax
90%Corporation Tax
95%Corporation Tax
95%Corporation Tax
85%Corporation Tax
90%Corporation Tax
95%Financial Statement
90%Financial Statement
90%Corporation Tax
95%Financial Statement
95%Corporation Tax
90%Corporation Tax
95%Corporation Tax
90%Unknown
70%Corporation Tax
95%Financial Statement
85%Unknown
80%P32 Payroll
95%P32 Payroll
95%Corporation Tax
90%Unknown
80%Financial Statement
95%Financial Statement
95%| Document Type | Pages | Confidence | Processing Status | Created | Actions |
|---|---|---|---|---|---|
| Corporation Tax | 1-2 (2 pages) | 95% | Processed | 2026-05-11 12:57 | No view available |
| Corporation Tax | 3-11 (9 pages) | 95% | Processed | 2026-05-11 12:57 | No view available |
| Financial Statement | 12 (1 page) | 95% | Processed | 2026-05-11 12:57 | No view available |
| Financial Statement | 13 (1 page) | 95% | Processed | 2026-05-11 12:57 | No view available |
| Corporation Tax | 14 (1 page) | 95% | Processed | 2026-05-11 12:57 | No view available |
| Financial Statement | 15 (1 page) | 95% | Processed | 2026-05-11 12:57 | No view available |
| Corporation Tax | 16 (1 page) | 95% | Processed | 2026-05-11 12:57 | No view available |
| Corporation Tax | 17 (1 page) | 95% | Processed | 2026-05-11 12:57 | No view available |
| Corporation Tax | 18 (1 page) | 90% | Processed | 2026-05-11 12:57 | No view available |
| Corporation Tax | 20 (1 page) | 95% | Processed | 2026-05-11 12:57 | No view available |
| Financial Statement | 21 (1 page) | 85% | Processed | 2026-05-11 12:57 | No view available |
| P32 Payroll | 23-24 (2 pages) | 95% | Processed | 2026-05-11 12:57 | No view available |
| Corporation Tax | 25 (1 page) | 90% | Processed | 2026-05-11 12:57 | No view available |
| Financial Statement | 27-28 (2 pages) | 95% | Processed | 2026-05-11 12:57 | No view available |
{
"document_groups": [
{
"avg_confidence": 0.9272727272727272,
"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 references to tax repayment claims and IRmark numbers, which are specific to 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-DeKT3NiK6kyna3Qlf7bS5V5Phcfva\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 references to tax repayment claims and IRmark numbers, which are specific to 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=1778504169, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=226, prompt_tokens=1383, total_tokens=1609, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_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 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 references to tax repayment claims and IRmark numbers, which are specific to tax filings.",
"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 references to tax repayment claims and IRmark numbers, which are specific to 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": 0.13270588235294117,
"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": 1885
},
{
"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 which are indicative of a corporation tax return or related 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-DeKT5Sg4cpjIMFmghL1oHg8nq9ucr\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 which are indicative of a corporation tax return or related 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=1778504171, 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=1036, total_tokens=1252, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptToke"
},
"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 which are indicative of a corporation tax return or related tax documentation.",
"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 which are indicative of a corporation tax return or related 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": 0.03333333333333333,
"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": 194
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a tax return for the year 2024-2025, UTR (Unique Taxpayer Reference), and instructions for filing a self-assessment tax return. 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-DeKT6ZwPXFh5RiAkcvvDgPd5wvvfk\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a tax return for the year 2024-2025, UTR (Unique Taxpayer Reference), and instructions for filing a self-assessment tax return. 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=1778504172, 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=1734, total_tokens=1950, 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, cach"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HM Revenue \u0026 Customs, a tax return for the year 2024-2025, UTR (Unique Taxpayer Reference), and instructions for filing a self-assessment tax return. These are key indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a tax return for the year 2024-2025, UTR (Unique Taxpayer Reference), and instructions for filing a self-assessment tax return. 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": 0.1807058823529412,
"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": 2695
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and UK property income. These are key indicators of a corporation tax document or a tax return form. The structure and content align with a tax-related document rather than any other category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKT79Pay9gFZ8BVVbAs6Cm2lAQYA\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and UK property income. These are key indicators of a corporation tax document or a tax return form. The structure and content align with a tax-related document rather than any other category.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504173, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=230, prompt_tokens=1726, total_tokens=1956, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, 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 a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and UK property income. These are key indicators of a corporation tax document or a tax return form. The structure and content align with a tax-related document rather than any other category.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and UK property income. These are key indicators of a corporation tax document or a tax return form. The structure and content align with a tax-related document rather than any other category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.0757894736842105,
"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": 4060
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027State Pension\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027Jobseeker\u0027s Allowance\u0027. Additionally, the document includes references to specific tax boxes and supplementary pages, which are indicative of a tax return document. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return form, likely related to corporation tax.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKT9HcNRmSRtTxe6C7hS897s3j7l\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Taxed UK interest\\\u0027, \\\u0027Untaxed UK interest\\\u0027, \\\u0027Foreign dividends\\\u0027, \\\u0027State Pension\\\u0027, \\\u0027Taxable Incapacity Benefit\\\u0027, and \\\u0027Jobseeker\\\u0027s Allowance\\\u0027. Additionally, the document includes references to specific tax boxes and supplementary pages, which are indicative of a tax return document. The presence of \\\u0027SA 100\\\u0027 further confirms that this is a self-assessment tax return form, likely related to corporation tax.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504175, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(compl"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027State Pension\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027Jobseeker\u0027s Allowance\u0027. Additionally, the document includes references to specific tax boxes and supplementary pages, which are indicative of a tax return document. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return form, likely related to corporation tax.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027State Pension\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027Jobseeker\u0027s Allowance\u0027. Additionally, the document includes references to specific tax boxes and supplementary pages, which are indicative of a tax return document. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return form, likely related to corporation tax.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 2232
},
{
"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 a tax reference number and references to specific tax years (e.g., \u0027year to 5 April 2025\u0027) strongly indicate that this is 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-DeKTAoryxcsMRtws6BqGBYpuCj0Rz\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 a tax reference number and references to specific tax years (e.g., \\\u0027year to 5 April 2025\\\u0027) strongly indicate that this is 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=1778504176, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=231, prompt_tokens=1604, total_tokens=1835, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, "
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number and references to specific tax years (e.g., \u0027year to 5 April 2025\u0027) strongly indicate that this is 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 reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number and references to specific tax years (e.g., \u0027year to 5 April 2025\u0027) strongly indicate that this is 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": 0.11666666666666667,
"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": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKTCBDueBNdURAVcuubbadTcMRsg\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Student Loan and Postgraduate Loan repayments\\\u0027, \\\u0027High Income Child Benefit Charge\\\u0027, and \\\u0027Marriage Allowance\\\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \\\u0027SA 100\\\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504178, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11454545454545456,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.09090909090909091,
"total_score": 0.11454545454545456
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.030303030303030304,
"total_score": 0.06681818181818183
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "loan_statement"
}
},
"page_number": 7,
"text_length": 2179
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax codes, and HMRC-related information, which are key indicators of corporation tax documents. It also includes a tax reference number and mentions Self Assessment Statements, which are specific to 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-DeKTD67wz9Zt4rbeesduCYVZhS9i8\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax codes, and HMRC-related information, which are key indicators of corporation tax documents. It also includes a tax reference number and mentions Self Assessment Statements, which are specific to 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=1778504179, 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=1680, total_tokens=1892, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=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 references to tax returns, tax calculations, tax codes, and HMRC-related information, which are key indicators of corporation tax documents. It also includes a tax reference number and mentions Self Assessment Statements, which are specific to tax filings.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax codes, and HMRC-related information, which are key indicators of corporation tax documents. It also includes a tax reference number and mentions Self Assessment Statements, which are specific to 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": 0.1392857142857143,
"features": {
"bank_statement": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.0,
"total_score": 0.10588235294117647
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.14285714285714285,
"total_score": 0.1392857142857143
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.030303030303030304,
"total_score": 0.05181818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 8,
"text_length": 2552
},
{
"confidence_score": 0.85,
"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, \u0027SA100\u0027, and \u0027Your tax adviser\u0027. Additionally, it includes details about a tax adviser, client reference, and other tax-related identifiers, which are indicative of corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKTFclcbkhLuKtSDO1BQA7eSkObB\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, \\\u0027SA100\\\u0027, and \\\u0027Your tax adviser\\\u0027. Additionally, it includes details about a tax adviser, client reference, and other tax-related identifiers, which are indicative of corporation tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504181, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=215, prompt_tokens=1181, total_tokens=1396, 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 tax-related information such as \u0027Tax Reference\u0027, \u0027SA100\u0027, and \u0027Your tax adviser\u0027. Additionally, it includes details about a tax adviser, client reference, and other tax-related identifiers, which are indicative of corporation tax documents.",
"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, \u0027SA100\u0027, and \u0027Your tax adviser\u0027. Additionally, it includes details about a tax adviser, client reference, and other tax-related identifiers, which are indicative of corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"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": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. 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-DeKTGOrXMXcvdxEfWSkX5RUGZvTci\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. 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=1778504182, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=198, prompt_tokens=1269, total_tokens=1467, 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=1152), latency_checkpoint={\u0027engine_tbt_ms\u0027: 14, \u0027en"
},
"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 (SA 100), tax reference numbers, and declarations related to tax filings. These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. 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": 0.041666666666666664,
"pattern_score": 0.0,
"total_score": 0.045
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 10,
"text_length": 1070
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and tax reporting, aligning with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKTJfzx4FD8RN3tIn67XdQIhOjRd\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 and tax reporting, aligning with corporation tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504185, 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=1579, total_tokens=1793, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(au"
},
"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 and tax reporting, aligning with corporation tax documents.",
"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 and tax reporting, aligning with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "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": 0.25,
"pattern_score": 0.0,
"total_score": 0.195
},
"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": 1874
}
],
"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": 13,
"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 related to business expenses, including categories like wages, rent, depreciation, and total expenses. These are key indicators of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, invoices, payroll, loans, or property-related details.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKTLDuCwzH8wMWPHvuyFQUomMeAF\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 related to business expenses, including categories like wages, rent, depreciation, and total expenses. These are key indicators of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, invoices, payroll, loans, or property-related details.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504187, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=221, prompt_tokens=1592, total_tokens=1813, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reaso"
},
"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 related to business expenses, including categories like wages, rent, depreciation, and total expenses. These are key indicators of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, invoices, payroll, loans, or property-related details.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data related to business expenses, including categories like wages, rent, depreciation, and total expenses. These are key indicators of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, invoices, payroll, loans, or property-related details.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 1500
},
{
"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 net profit, capital allowances, taxable profit or loss, and other financial metrics. These are key indicators of a financial statement, specifically related to business income and expenses.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKTNcDb6jmqN2x51raEGiIXPuqm2\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 net profit, capital allowances, taxable profit or loss, and other financial metrics. These are key indicators of a financial statement, specifically related to business income and expenses.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504189, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=204, prompt_tokens=1725, total_tokens=1929, 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,"
},
"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 net profit, capital allowances, taxable profit or loss, and other financial metrics. These are key indicators of a financial statement, specifically related to business income and expenses.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data such as net profit, capital allowances, taxable profit or loss, and other financial metrics. These are key indicators of a financial statement, specifically related to business income and expenses.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 0.0,
"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": 2542
}
],
"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 taxable profits, losses, adjustments, and self-employment tax notes, which are key indicators of a corporation tax-related document. Specific references to tax years (e.g., 2024-25), taxable profits, and loss carry-forwards 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-DeKTO4bgD0jsva8gjwKi22rQJLoOC\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, losses, adjustments, and self-employment tax notes, which are key indicators of a corporation tax-related document. Specific references to tax years (e.g., 2024-25), taxable profits, and loss carry-forwards 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=1778504190, 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=1732, total_tokens=1954, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_"
},
"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, losses, adjustments, and self-employment tax notes, which are key indicators of a corporation tax-related document. Specific references to tax years (e.g., 2024-25), taxable profits, and loss carry-forwards further confirm this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profits, losses, adjustments, and self-employment tax notes, which are key indicators of a corporation tax-related document. Specific references to tax years (e.g., 2024-25), taxable profits, and loss carry-forwards 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": "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": 2603
}
],
"start_page": 14
},
{
"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 accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial 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-DeKTPJncMO1rd4JfTcPTUVS8DgaQp\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 accounts. Key indicators include terms like \\\u0027Balance sheet,\\\u0027 \\\u0027Assets,\\\u0027 \\\u0027Liabilities,\\\u0027 \\\u0027Net business assets,\\\u0027 and \\\u0027Capital account,\\\u0027 which are characteristic of financial 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=1778504191, 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=1550, total_tokens=1764, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), 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 with detailed information about assets, liabilities, and capital accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial statements.",
"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 accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial 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": 0.2374736842105263,
"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": 0.9166666666666666,
"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": 18,
"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 a tax reference number, National Insurance contributions (NICs), and self-employment profits. It also mentions specific tax-related terms such as \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 and \u0027adjustment to profits chargeable to Class 4 NICs,\u0027 which are indicative of a tax document. Additionally, the presence of a client reference and the form identifier \u0027SA103F\u0027 suggests it is related to a self-assessment tax return, 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-DeKTRfcUQeirVxKy6W6fzHydPr9oA\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 a tax reference number, National Insurance contributions (NICs), and self-employment profits. It also mentions specific tax-related terms such as \\\u0027Class 2 NICs,\\\u0027 \\\u0027Class 4 NICs,\\\u0027 and \\\u0027adjustment to profits chargeable to Class 4 NICs,\\\u0027 which are indicative of a tax document. Additionally, the presence of a client reference and the form identifier \\\u0027SA103F\\\u0027 suggests it is related to a self-assessment tax return, 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=1778504193, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and self-employment profits. It also mentions specific tax-related terms such as \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 and \u0027adjustment to profits chargeable to Class 4 NICs,\u0027 which are indicative of a tax document. Additionally, the presence of a client reference and the form identifier \u0027SA103F\u0027 suggests it is related to a self-assessment tax return, which aligns with corporation tax documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and self-employment profits. It also mentions specific tax-related terms such as \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 and \u0027adjustment to profits chargeable to Class 4 NICs,\u0027 which are indicative of a tax document. Additionally, the presence of a client reference and the form identifier \u0027SA103F\u0027 suggests it is related to a self-assessment tax return, 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": "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": 866
},
{
"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 Assessment, Unique Taxpayer Reference (UTR), and tax calculations, which are key indicators of a corporation tax document. It also includes details about payments on account, underpaid tax, and other debts, which are typical elements of tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKTTWrNifn5AnMCXNt502RGx6F81\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 Assessment, Unique Taxpayer Reference (UTR), and tax calculations, which are key indicators of a corporation tax document. It also includes details about payments on account, underpaid tax, and other debts, which are typical elements of tax-related documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504195, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=226, prompt_tokens=1705, total_tokens=1931, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_predictio"
},
"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 Assessment, Unique Taxpayer Reference (UTR), and tax calculations, which are key indicators of a corporation tax document. It also includes details about payments on account, underpaid tax, and other debts, which are typical elements of tax-related documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, Self Assessment, Unique Taxpayer Reference (UTR), and tax calculations, which are key indicators of a corporation tax document. It also includes details about payments on account, underpaid tax, and other debts, which are typical elements of tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.17505882352941174,
"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": 2299
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027Tax Reference,\u0027 \u0027adjustments to tax due,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKTV1v8QlvqxZQ4f65zu2Bt9bePe\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \\\u0027Tax Reference,\\\u0027 \\\u0027adjustments to tax due,\\\u0027 and \\\u0027self-assessment statement of account.\\\u0027 These are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504197, 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=1335, total_tokens=1562, 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 tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027Tax Reference,\u0027 \u0027adjustments to tax due,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027Tax Reference,\u0027 \u0027adjustments to tax due,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.07058823529411765,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "bank_statement"
}
},
"page_number": 18,
"text_length": 1363
}
],
"start_page": 16
},
{
"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": 20,
"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 tax-related information, including a computation summary, income tax details, National Insurance Contributions (NIC), and a tax reference number. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Taxable Income,\u0027 \u0027Income Tax Due,\u0027 and \u0027NIC\u0027 strongly suggests it is related to corporation tax.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKTZMvLbdk1EIzilT4iTQi9qzQbR\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, income tax details, National Insurance Contributions (NIC), and a tax reference number. These are key indicators of a tax return or corporation tax document. The presence of terms like \\\u0027Tax Return Schedules,\\\u0027 \\\u0027Taxable Income,\\\u0027 \\\u0027Income Tax Due,\\\u0027 and \\\u0027NIC\\\u0027 strongly suggests it is related to corporation tax.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504201, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=238, prompt_tokens=1293, total_tokens=1531, completion_tokens_details=CompletionTokensDetails(accepte"
},
"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, income tax details, National Insurance Contributions (NIC), and a tax reference number. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Taxable Income,\u0027 \u0027Income Tax Due,\u0027 and \u0027NIC\u0027 strongly suggests it is related to corporation tax.",
"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, income tax details, National Insurance Contributions (NIC), and a tax reference number. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Taxable Income,\u0027 \u0027Income Tax Due,\u0027 and \u0027NIC\u0027 strongly suggests it is related to corporation tax.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 0.0,
"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": 0.14285714285714285,
"total_score": 0.1392857142857143
},
"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": "nominal_ledger"
}
},
"page_number": 20,
"text_length": 888
}
],
"start_page": 20
},
{
"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. It does not include transaction details, 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-DeKTaIE7O0dCagAkL134s7QPAFfgw\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. It does not include transaction details, 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=1778504202, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=215, prompt_tokens=1974, total_tokens=2189, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), pro"
},
"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. It does not include transaction details, 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. It does not include transaction details, 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": 21,
"text_length": 1718
}
],
"start_page": 21
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "p32_payroll",
"end_page": 24,
"pages": [
{
"confidence_score": 0.95,
"document_type": "p32_payroll",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure aligns with tax obligations and payment schedules, which are characteristic of P32 payroll 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-DeKTeKQJF5C3IO6h5bPTUjdgLovHc\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"p32_payroll\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to \\\u0027Client Payment Advice,\\\u0027 \\\u0027Payments Required,\\\u0027 \\\u0027Payment Dates,\\\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure aligns with tax obligations and payment schedules, which are characteristic of P32 payroll 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=1778504206, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=220, prompt_tokens=1146, total_tokens=1366, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure aligns with tax obligations and payment schedules, which are characteristic of P32 payroll documents.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure aligns with tax obligations and payment schedules, which are characteristic of P32 payroll 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": "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": 23,
"text_length": 412
},
{
"confidence_score": 0.95,
"document_type": "p32_payroll",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax obligations, Class 4 NIC (National Insurance Contributions), and payment on account instalments for a specific tax year, which are key indicators of an HMRC P32 Employer Payment Record.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKTfNpDqa3LGdmRbSCNyTy8ktDUG\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"p32_payroll\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax obligations, Class 4 NIC (National Insurance Contributions), and payment on account instalments for a specific tax year, which are key indicators of an HMRC P32 Employer Payment Record.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504207, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=208, prompt_tokens=1130, total_tokens=1338, 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=1024), lat"
},
"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 obligations, Class 4 NIC (National Insurance Contributions), and payment on account instalments for a specific tax year, which are key indicators of an HMRC P32 Employer Payment Record.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax obligations, Class 4 NIC (National Insurance Contributions), and payment on account instalments for a specific tax year, which are key indicators of an HMRC P32 Employer Payment Record.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 0.013888888888888888,
"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": 24,
"text_length": 397
}
],
"start_page": 23
},
{
"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": 25,
"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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of terms like \u0027Class 4 NIC\u0027, \u0027Self-Employment Profit\u0027, and \u0027Contribution\u0027 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-DeKTjnfwhkumd62sh20eBVgeSFwcw\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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of terms like \\\u0027Class 4 NIC\\\u0027, \\\u0027Self-Employment Profit\\\u0027, and \\\u0027Contribution\\\u0027 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=1778504211, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=229, prompt_tokens=1122, total_tokens=1351, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of terms like \u0027Class 4 NIC\u0027, \u0027Self-Employment Profit\u0027, and \u0027Contribution\u0027 aligns with corporation tax or self-assessment tax documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of terms like \u0027Class 4 NIC\u0027, \u0027Self-Employment Profit\u0027, and \u0027Contribution\u0027 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": 25,
"text_length": 361
}
],
"start_page": 25
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "financial_statement",
"end_page": 28,
"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 detailed financial data such as sales income, direct costs, employee costs, and net profit. These are key indicators of a financial statement, specifically an income statement or profit and loss statement. There are no indicators of transactions, invoices, payroll, loans, mortgages, or property-related details.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKTm38Gl1PXx3PrzWroO4h2rdcA1\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 detailed financial data such as sales income, direct costs, employee costs, and net profit. These are key indicators of a financial statement, specifically an income statement or profit and loss statement. There are no indicators of transactions, invoices, payroll, loans, mortgages, or property-related details.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504214, 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=1373, total_tokens=1606, completion_tokens_details=CompletionTokensDetails(accepted_predi"
},
"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 detailed financial data such as sales income, direct costs, employee costs, and net profit. These are key indicators of a financial statement, specifically an income statement or profit and loss statement. There are no indicators of transactions, invoices, payroll, loans, mortgages, or property-related details.",
"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 detailed financial data such as sales income, direct costs, employee costs, and net profit. These are key indicators of a financial statement, specifically an income statement or profit and loss statement. There are no indicators of transactions, invoices, payroll, loans, mortgages, or property-related details.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 0.16666666666666666,
"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": 0.013888888888888888,
"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": 1214
},
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes net profit calculations, adjustments, and tax-related profit figures, 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-DeKToWHGuSVEe7gKYmkDcXiXnegHI\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes net profit calculations, adjustments, and tax-related profit figures, 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=1778504216, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=207, prompt_tokens=1337, total_tokens=1544, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(aud"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes net profit calculations, adjustments, and tax-related profit figures, 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 capital accounts. It also includes net profit calculations, adjustments, and tax-related profit figures, 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": 0.295578947368421,
"features": {
"bank_statement": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.03529411764705882
},
"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": 0.21052631578947367,
"pattern_score": 0.3,
"total_score": 0.295578947368421
},
"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": 0.013888888888888888,
"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": 28,
"text_length": 1028
}
],
"start_page": 27
}
],
"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 references to tax repayment claims and IRmark numbers, which are specific to 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-DeKT3NiK6kyna3Qlf7bS5V5Phcfva\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 references to tax repayment claims and IRmark numbers, which are specific to 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=1778504169, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=226, prompt_tokens=1383, total_tokens=1609, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_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 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 references to tax repayment claims and IRmark numbers, which are specific to tax filings.",
"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 references to tax repayment claims and IRmark numbers, which are specific to 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": 0.13270588235294117,
"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": 1885
},
{
"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 which are indicative of a corporation tax return or related 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-DeKT5Sg4cpjIMFmghL1oHg8nq9ucr\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 which are indicative of a corporation tax return or related 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=1778504171, 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=1036, total_tokens=1252, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptToke"
},
"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 which are indicative of a corporation tax return or related tax documentation.",
"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 which are indicative of a corporation tax return or related 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": 0.03333333333333333,
"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": 194
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a tax return for the year 2024-2025, UTR (Unique Taxpayer Reference), and instructions for filing a self-assessment tax return. 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-DeKT6ZwPXFh5RiAkcvvDgPd5wvvfk\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a tax return for the year 2024-2025, UTR (Unique Taxpayer Reference), and instructions for filing a self-assessment tax return. 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=1778504172, 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=1734, total_tokens=1950, 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, cach"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HM Revenue \u0026 Customs, a tax return for the year 2024-2025, UTR (Unique Taxpayer Reference), and instructions for filing a self-assessment tax return. These are key indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a tax return for the year 2024-2025, UTR (Unique Taxpayer Reference), and instructions for filing a self-assessment tax return. 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": 0.1807058823529412,
"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": 2695
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and UK property income. These are key indicators of a corporation tax document or a tax return form. The structure and content align with a tax-related document rather than any other category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKT79Pay9gFZ8BVVbAs6Cm2lAQYA\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and UK property income. These are key indicators of a corporation tax document or a tax return form. The structure and content align with a tax-related document rather than any other category.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504173, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=230, prompt_tokens=1726, total_tokens=1956, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, 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 a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and UK property income. These are key indicators of a corporation tax document or a tax return form. The structure and content align with a tax-related document rather than any other category.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and UK property income. These are key indicators of a corporation tax document or a tax return form. The structure and content align with a tax-related document rather than any other category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.0757894736842105,
"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": 4060
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027State Pension\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027Jobseeker\u0027s Allowance\u0027. Additionally, the document includes references to specific tax boxes and supplementary pages, which are indicative of a tax return document. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return form, likely related to corporation tax.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKT9HcNRmSRtTxe6C7hS897s3j7l\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Taxed UK interest\\\u0027, \\\u0027Untaxed UK interest\\\u0027, \\\u0027Foreign dividends\\\u0027, \\\u0027State Pension\\\u0027, \\\u0027Taxable Incapacity Benefit\\\u0027, and \\\u0027Jobseeker\\\u0027s Allowance\\\u0027. Additionally, the document includes references to specific tax boxes and supplementary pages, which are indicative of a tax return document. The presence of \\\u0027SA 100\\\u0027 further confirms that this is a self-assessment tax return form, likely related to corporation tax.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504175, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(compl"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027State Pension\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027Jobseeker\u0027s Allowance\u0027. Additionally, the document includes references to specific tax boxes and supplementary pages, which are indicative of a tax return document. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return form, likely related to corporation tax.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027State Pension\u0027, \u0027Taxable Incapacity Benefit\u0027, and \u0027Jobseeker\u0027s Allowance\u0027. Additionally, the document includes references to specific tax boxes and supplementary pages, which are indicative of a tax return document. The presence of \u0027SA 100\u0027 further confirms that this is a self-assessment tax return form, likely related to corporation tax.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 2232
},
{
"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 a tax reference number and references to specific tax years (e.g., \u0027year to 5 April 2025\u0027) strongly indicate that this is 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-DeKTAoryxcsMRtws6BqGBYpuCj0Rz\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 a tax reference number and references to specific tax years (e.g., \\\u0027year to 5 April 2025\\\u0027) strongly indicate that this is 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=1778504176, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=231, prompt_tokens=1604, total_tokens=1835, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, "
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number and references to specific tax years (e.g., \u0027year to 5 April 2025\u0027) strongly indicate that this is 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 reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number and references to specific tax years (e.g., \u0027year to 5 April 2025\u0027) strongly indicate that this is 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": 0.11666666666666667,
"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": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKTCBDueBNdURAVcuubbadTcMRsg\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Student Loan and Postgraduate Loan repayments\\\u0027, \\\u0027High Income Child Benefit Charge\\\u0027, and \\\u0027Marriage Allowance\\\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \\\u0027SA 100\\\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504178, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11454545454545456,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.09090909090909091,
"total_score": 0.11454545454545456
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.030303030303030304,
"total_score": 0.06681818181818183
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "loan_statement"
}
},
"page_number": 7,
"text_length": 2179
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax codes, and HMRC-related information, which are key indicators of corporation tax documents. It also includes a tax reference number and mentions Self Assessment Statements, which are specific to 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-DeKTD67wz9Zt4rbeesduCYVZhS9i8\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax codes, and HMRC-related information, which are key indicators of corporation tax documents. It also includes a tax reference number and mentions Self Assessment Statements, which are specific to 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=1778504179, 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=1680, total_tokens=1892, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=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 references to tax returns, tax calculations, tax codes, and HMRC-related information, which are key indicators of corporation tax documents. It also includes a tax reference number and mentions Self Assessment Statements, which are specific to tax filings.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, tax codes, and HMRC-related information, which are key indicators of corporation tax documents. It also includes a tax reference number and mentions Self Assessment Statements, which are specific to 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": 0.1392857142857143,
"features": {
"bank_statement": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.0,
"total_score": 0.10588235294117647
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.14285714285714285,
"total_score": 0.1392857142857143
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.030303030303030304,
"total_score": 0.05181818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 8,
"text_length": 2552
},
{
"confidence_score": 0.85,
"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, \u0027SA100\u0027, and \u0027Your tax adviser\u0027. Additionally, it includes details about a tax adviser, client reference, and other tax-related identifiers, which are indicative of corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKTFclcbkhLuKtSDO1BQA7eSkObB\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, \\\u0027SA100\\\u0027, and \\\u0027Your tax adviser\\\u0027. Additionally, it includes details about a tax adviser, client reference, and other tax-related identifiers, which are indicative of corporation tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504181, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=215, prompt_tokens=1181, total_tokens=1396, 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 tax-related information such as \u0027Tax Reference\u0027, \u0027SA100\u0027, and \u0027Your tax adviser\u0027. Additionally, it includes details about a tax adviser, client reference, and other tax-related identifiers, which are indicative of corporation tax documents.",
"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, \u0027SA100\u0027, and \u0027Your tax adviser\u0027. Additionally, it includes details about a tax adviser, client reference, and other tax-related identifiers, which are indicative of corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"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": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. 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-DeKTGOrXMXcvdxEfWSkX5RUGZvTci\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. 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=1778504182, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=198, prompt_tokens=1269, total_tokens=1467, 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=1152), latency_checkpoint={\u0027engine_tbt_ms\u0027: 14, \u0027en"
},
"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 (SA 100), tax reference numbers, and declarations related to tax filings. These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. 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": 0.041666666666666664,
"pattern_score": 0.0,
"total_score": 0.045
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 10,
"text_length": 1070
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and tax reporting, aligning with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKTJfzx4FD8RN3tIn67XdQIhOjRd\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 and tax reporting, aligning with corporation tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504185, 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=1579, total_tokens=1793, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(au"
},
"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 and tax reporting, aligning with corporation tax documents.",
"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 and tax reporting, aligning with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "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": 0.25,
"pattern_score": 0.0,
"total_score": 0.195
},
"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": 1874
},
{
"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 related to business expenses, including categories like wages, rent, depreciation, and total expenses. These are key indicators of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, invoices, payroll, loans, or property-related details.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKTLDuCwzH8wMWPHvuyFQUomMeAF\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 related to business expenses, including categories like wages, rent, depreciation, and total expenses. These are key indicators of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, invoices, payroll, loans, or property-related details.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504187, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=221, prompt_tokens=1592, total_tokens=1813, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reaso"
},
"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 related to business expenses, including categories like wages, rent, depreciation, and total expenses. These are key indicators of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, invoices, payroll, loans, or property-related details.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data related to business expenses, including categories like wages, rent, depreciation, and total expenses. These are key indicators of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, invoices, payroll, loans, or property-related details.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 1500
},
{
"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 net profit, capital allowances, taxable profit or loss, and other financial metrics. These are key indicators of a financial statement, specifically related to business income and expenses.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKTNcDb6jmqN2x51raEGiIXPuqm2\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 net profit, capital allowances, taxable profit or loss, and other financial metrics. These are key indicators of a financial statement, specifically related to business income and expenses.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504189, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=204, prompt_tokens=1725, total_tokens=1929, 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,"
},
"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 net profit, capital allowances, taxable profit or loss, and other financial metrics. These are key indicators of a financial statement, specifically related to business income and expenses.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data such as net profit, capital allowances, taxable profit or loss, and other financial metrics. These are key indicators of a financial statement, specifically related to business income and expenses.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 0.0,
"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": 2542
},
{
"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, losses, adjustments, and self-employment tax notes, which are key indicators of a corporation tax-related document. Specific references to tax years (e.g., 2024-25), taxable profits, and loss carry-forwards 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-DeKTO4bgD0jsva8gjwKi22rQJLoOC\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, losses, adjustments, and self-employment tax notes, which are key indicators of a corporation tax-related document. Specific references to tax years (e.g., 2024-25), taxable profits, and loss carry-forwards 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=1778504190, 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=1732, total_tokens=1954, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_"
},
"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, losses, adjustments, and self-employment tax notes, which are key indicators of a corporation tax-related document. Specific references to tax years (e.g., 2024-25), taxable profits, and loss carry-forwards further confirm this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profits, losses, adjustments, and self-employment tax notes, which are key indicators of a corporation tax-related document. Specific references to tax years (e.g., 2024-25), taxable profits, and loss carry-forwards 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": "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": 2603
},
{
"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 accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial 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-DeKTPJncMO1rd4JfTcPTUVS8DgaQp\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 accounts. Key indicators include terms like \\\u0027Balance sheet,\\\u0027 \\\u0027Assets,\\\u0027 \\\u0027Liabilities,\\\u0027 \\\u0027Net business assets,\\\u0027 and \\\u0027Capital account,\\\u0027 which are characteristic of financial 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=1778504191, 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=1550, total_tokens=1764, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), 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 with detailed information about assets, liabilities, and capital accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial statements.",
"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 accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial 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": 0.2374736842105263,
"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 a tax reference number, National Insurance contributions (NICs), and self-employment profits. It also mentions specific tax-related terms such as \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 and \u0027adjustment to profits chargeable to Class 4 NICs,\u0027 which are indicative of a tax document. Additionally, the presence of a client reference and the form identifier \u0027SA103F\u0027 suggests it is related to a self-assessment tax return, 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-DeKTRfcUQeirVxKy6W6fzHydPr9oA\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 a tax reference number, National Insurance contributions (NICs), and self-employment profits. It also mentions specific tax-related terms such as \\\u0027Class 2 NICs,\\\u0027 \\\u0027Class 4 NICs,\\\u0027 and \\\u0027adjustment to profits chargeable to Class 4 NICs,\\\u0027 which are indicative of a tax document. Additionally, the presence of a client reference and the form identifier \\\u0027SA103F\\\u0027 suggests it is related to a self-assessment tax return, 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=1778504193, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and self-employment profits. It also mentions specific tax-related terms such as \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 and \u0027adjustment to profits chargeable to Class 4 NICs,\u0027 which are indicative of a tax document. Additionally, the presence of a client reference and the form identifier \u0027SA103F\u0027 suggests it is related to a self-assessment tax return, which aligns with corporation tax documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and self-employment profits. It also mentions specific tax-related terms such as \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 and \u0027adjustment to profits chargeable to Class 4 NICs,\u0027 which are indicative of a tax document. Additionally, the presence of a client reference and the form identifier \u0027SA103F\u0027 suggests it is related to a self-assessment tax return, 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": "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": 866
},
{
"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 Assessment, Unique Taxpayer Reference (UTR), and tax calculations, which are key indicators of a corporation tax document. It also includes details about payments on account, underpaid tax, and other debts, which are typical elements of tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKTTWrNifn5AnMCXNt502RGx6F81\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 Assessment, Unique Taxpayer Reference (UTR), and tax calculations, which are key indicators of a corporation tax document. It also includes details about payments on account, underpaid tax, and other debts, which are typical elements of tax-related documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504195, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=226, prompt_tokens=1705, total_tokens=1931, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_predictio"
},
"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 Assessment, Unique Taxpayer Reference (UTR), and tax calculations, which are key indicators of a corporation tax document. It also includes details about payments on account, underpaid tax, and other debts, which are typical elements of tax-related documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, Self Assessment, Unique Taxpayer Reference (UTR), and tax calculations, which are key indicators of a corporation tax document. It also includes details about payments on account, underpaid tax, and other debts, which are typical elements of tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.17505882352941174,
"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": 2299
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027Tax Reference,\u0027 \u0027adjustments to tax due,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKTV1v8QlvqxZQ4f65zu2Bt9bePe\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \\\u0027Tax Reference,\\\u0027 \\\u0027adjustments to tax due,\\\u0027 and \\\u0027self-assessment statement of account.\\\u0027 These are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778504197, 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=1335, total_tokens=1562, 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 tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027Tax Reference,\u0027 \u0027adjustments to tax due,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027Tax Reference,\u0027 \u0027adjustments to tax due,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.07058823529411765,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "bank_statement"
}
},
"page_number": 18,
"text_length": 1363
},
{
"confidence_score": 0.7,
"document_type": "unknown",
"features": {
"llm_based": {
"confidence": 0.7,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.7,\n \"reasoning\": \"The document mentions \u0027Self-employment\u0027, \u0027business name\u0027, and \u0027APE (Accounting Period End)\u0027, which are indicative of financial reporting or tax-related documentation. However, it does not explicitly match the structure or content of other categories like invoices, payroll, or loan statements. The lack of specific identifiers for other categories leads to the classification as 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-DeKTXNC0ej7RhTEwgsXsddGK39xd9\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.7,\\n \"reasoning\": \"The document mentions \\\u0027Self-employment\\\u0027, \\\u0027business name\\\u0027, and \\\u0027APE (Accounting Period End)\\\u0027, which are indicative of financial reporting or tax-related documentation. However, it does not explicitly match the structure or content of other categories like invoices, payroll, or loan statements. The lack of specific identifiers for other categories leads to the classification as 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=1778504199, 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=1055, total_tokens=1288, 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 mentions \u0027Self-employment\u0027, \u0027business name\u0027, and \u0027APE (Accounting Period End)\u0027, which are indicative of financial reporting or tax-related documentation. However, it does not explicitly match the structure or content of other categories like invoices, payroll, or loan statements. The lack of specific identifiers for other categories leads to the classification as a financial statement.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.7,\n \"reasoning\": \"The document mentions \u0027Self-employment\u0027, \u0027business name\u0027, and \u0027APE (Accounting Period End)\u0027, which are indicative of financial reporting or tax-related documentation. However, it does not explicitly match the structure or content of other categories like invoices, payroll, or loan statements. The lack of specific identifiers for other categories leads to the classification as 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": "uncertain",
"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": 0.0,
"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": 0.08333333333333333,
"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": 0.0,
"total_score": 0.0
}
},
"type": "nominal_ledger"
}
},
"page_number": 19,
"text_length": 220
},
{
"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, income tax details, National Insurance Contributions (NIC), and a tax reference number. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Taxable Income,\u0027 \u0027Income Tax Due,\u0027 and \u0027NIC\u0027 strongly suggests it is related to corporation tax.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKTZMvLbdk1EIzilT4iTQi9qzQbR\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, income tax details, National Insurance Contributions (NIC), and a tax reference number. These are key indicators of a tax return or corporation tax document. The presence of terms like \\\u0027Tax Return Schedules,\\\u0027 \\\u0027Taxable Income,\\\u0027 \\\u0027Income Tax Due,\\\u0027 and \\\u0027NIC\\\u0027 strongly suggests it is related to corporation tax.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504201, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=238, prompt_tokens=1293, total_tokens=1531, completion_tokens_details=CompletionTokensDetails(accepte"
},
"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, income tax details, National Insurance Contributions (NIC), and a tax reference number. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Taxable Income,\u0027 \u0027Income Tax Due,\u0027 and \u0027NIC\u0027 strongly suggests it is related to corporation tax.",
"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, income tax details, National Insurance Contributions (NIC), and a tax reference number. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Taxable Income,\u0027 \u0027Income Tax Due,\u0027 and \u0027NIC\u0027 strongly suggests it is related to corporation tax.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 0.0,
"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": 0.14285714285714285,
"total_score": 0.1392857142857143
},
"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": "nominal_ledger"
}
},
"page_number": 20,
"text_length": 888
},
{
"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. It does not include transaction details, 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-DeKTaIE7O0dCagAkL134s7QPAFfgw\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. It does not include transaction details, 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=1778504202, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=215, prompt_tokens=1974, total_tokens=2189, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), pro"
},
"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. It does not include transaction details, 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. It does not include transaction details, 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": 21,
"text_length": 1718
},
{
"confidence_score": 0.8,
"document_type": "unknown",
"features": {
"llm_based": {
"confidence": 0.8,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.8,\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, and other financial metrics, which are typically found in financial statements or tax-related summaries. However, it does not match the structure of corporation tax returns or other specific tax forms like CT600. The lack of transactional details, account numbers, or property-specific information 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-DeKTcxTnTjetPPfT9vNp5OhQcZEE7\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.8,\\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, and other financial metrics, which are typically found in financial statements or tax-related summaries. However, it does not match the structure of corporation tax returns or other specific tax forms like CT600. The lack of transactional details, account numbers, or property-specific information 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=1778504204, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=236, prompt_tokens=1115, total_tokens=1351, completion_tokens_detail"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to capital gains tax (CGT), business asset disposal relief gains, and other financial metrics, which are typically found in financial statements or tax-related summaries. However, it does not match the structure of corporation tax returns or other specific tax forms like CT600. The lack of transactional details, account numbers, or property-specific information rules out other categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.8,\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, and other financial metrics, which are typically found in financial statements or tax-related summaries. However, it does not match the structure of corporation tax returns or other specific tax forms like CT600. The lack of transactional details, account numbers, or property-specific information 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": "uncertain",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 22,
"text_length": 394
},
{
"confidence_score": 0.95,
"document_type": "p32_payroll",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure aligns with tax obligations and payment schedules, which are characteristic of P32 payroll 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-DeKTeKQJF5C3IO6h5bPTUjdgLovHc\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"p32_payroll\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to \\\u0027Client Payment Advice,\\\u0027 \\\u0027Payments Required,\\\u0027 \\\u0027Payment Dates,\\\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure aligns with tax obligations and payment schedules, which are characteristic of P32 payroll 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=1778504206, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=220, prompt_tokens=1146, total_tokens=1366, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure aligns with tax obligations and payment schedules, which are characteristic of P32 payroll documents.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure aligns with tax obligations and payment schedules, which are characteristic of P32 payroll 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": "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": 23,
"text_length": 412
},
{
"confidence_score": 0.95,
"document_type": "p32_payroll",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax obligations, Class 4 NIC (National Insurance Contributions), and payment on account instalments for a specific tax year, which are key indicators of an HMRC P32 Employer Payment Record.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKTfNpDqa3LGdmRbSCNyTy8ktDUG\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"p32_payroll\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax obligations, Class 4 NIC (National Insurance Contributions), and payment on account instalments for a specific tax year, which are key indicators of an HMRC P32 Employer Payment Record.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504207, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=208, prompt_tokens=1130, total_tokens=1338, 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=1024), lat"
},
"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 obligations, Class 4 NIC (National Insurance Contributions), and payment on account instalments for a specific tax year, which are key indicators of an HMRC P32 Employer Payment Record.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax obligations, Class 4 NIC (National Insurance Contributions), and payment on account instalments for a specific tax year, which are key indicators of an HMRC P32 Employer Payment Record.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 0.013888888888888888,
"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": 24,
"text_length": 397
},
{
"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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of terms like \u0027Class 4 NIC\u0027, \u0027Self-Employment Profit\u0027, and \u0027Contribution\u0027 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-DeKTjnfwhkumd62sh20eBVgeSFwcw\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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of terms like \\\u0027Class 4 NIC\\\u0027, \\\u0027Self-Employment Profit\\\u0027, and \\\u0027Contribution\\\u0027 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=1778504211, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=229, prompt_tokens=1122, total_tokens=1351, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_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 Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of terms like \u0027Class 4 NIC\u0027, \u0027Self-Employment Profit\u0027, and \u0027Contribution\u0027 aligns with corporation tax or self-assessment tax documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to Class 4 National Insurance Contributions (NIC), self-employment profit, and tax-related calculations, which are indicative of a tax-related document. The presence of terms like \u0027Class 4 NIC\u0027, \u0027Self-Employment Profit\u0027, and \u0027Contribution\u0027 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": 25,
"text_length": 361
},
{
"confidence_score": 0.8,
"document_type": "unknown",
"features": {
"llm_based": {
"confidence": 0.8,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.8,\n \"reasoning\": \"The document mentions \u0027Year ended 5 April 2025\u0027 and \u0027Personal Allowances,\u0027 which are indicative of a financial summary or statement. However, it lacks specific details to classify it as a tax return, payroll statement, or other categories. 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-DeKTkmqQZRszJIARG5RqXHKrIEb3B\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.8,\\n \"reasoning\": \"The document mentions \\\u0027Year ended 5 April 2025\\\u0027 and \\\u0027Personal Allowances,\\\u0027 which are indicative of a financial summary or statement. However, it lacks specific details to classify it as a tax return, payroll statement, or other categories. 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=1778504212, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=230, prompt_tokens=1056, total_tokens=1286, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, au"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document mentions \u0027Year ended 5 April 2025\u0027 and \u0027Personal Allowances,\u0027 which are indicative of a financial summary or statement. However, it lacks specific details to classify it as a tax return, payroll statement, or other categories. 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 mentions \u0027Year ended 5 April 2025\u0027 and \u0027Personal Allowances,\u0027 which are indicative of a financial summary or statement. However, it lacks specific details to classify it as a tax return, payroll statement, or other categories. 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": 26,
"text_length": 178
},
{
"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 detailed financial data such as sales income, direct costs, employee costs, and net profit. These are key indicators of a financial statement, specifically an income statement or profit and loss statement. There are no indicators of transactions, invoices, payroll, loans, mortgages, or property-related details.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKTm38Gl1PXx3PrzWroO4h2rdcA1\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 detailed financial data such as sales income, direct costs, employee costs, and net profit. These are key indicators of a financial statement, specifically an income statement or profit and loss statement. There are no indicators of transactions, invoices, payroll, loans, mortgages, or property-related details.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778504214, 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=1373, total_tokens=1606, completion_tokens_details=CompletionTokensDetails(accepted_predi"
},
"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 detailed financial data such as sales income, direct costs, employee costs, and net profit. These are key indicators of a financial statement, specifically an income statement or profit and loss statement. There are no indicators of transactions, invoices, payroll, loans, mortgages, or property-related details.",
"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 detailed financial data such as sales income, direct costs, employee costs, and net profit. These are key indicators of a financial statement, specifically an income statement or profit and loss statement. There are no indicators of transactions, invoices, payroll, loans, mortgages, or property-related details.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 0.16666666666666666,
"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": 0.013888888888888888,
"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": 1214
},
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes net profit calculations, adjustments, and tax-related profit figures, 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-DeKToWHGuSVEe7gKYmkDcXiXnegHI\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes net profit calculations, adjustments, and tax-related profit figures, 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=1778504216, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=207, prompt_tokens=1337, total_tokens=1544, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(aud"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a balance sheet summary with detailed breakdowns of assets, liabilities, and capital accounts. It also includes net profit calculations, adjustments, and tax-related profit figures, 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 capital accounts. It also includes net profit calculations, adjustments, and tax-related profit figures, 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": 0.295578947368421,
"features": {
"bank_statement": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.03529411764705882
},
"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": 0.21052631578947367,
"pattern_score": 0.3,
"total_score": 0.295578947368421
},
"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": 0.013888888888888888,
"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": 28,
"text_length": 1028
}
],
"processing_timestamp": "2026-05-11T12:56:57.506264"
}