| Original Filename: | 1THO007_Thomas_M_SA100_24-25_1778503609.pdf |
| Upload Date: | May 11, 2026 at 12:51 PM |
| Total Pages: | 28 |
| Processing Status: | Completed |
| Total Documents: | 14 |
| Corporation Taxs: | 9 |
| Financial Statements: | 4 |
| P32 Payrolls: | 1 |
Corporation Tax
95%Corporation Tax
90%Corporation Tax
95%Corporation Tax
95%Corporation Tax
95%Corporation Tax
95%Corporation Tax
95%Corporation Tax
95%Corporation Tax
95%Corporation Tax
85%Corporation Tax
95%Financial Statement
85%Corporation Tax
95%Corporation Tax
95%Financial Statement
90%Corporation Tax
90%Corporation Tax
95%Corporation Tax
90%Unknown
70%Corporation Tax
95%Financial Statement
90%Unknown
80%P32 Payroll
95%Corporation Tax
95%Corporation Tax
90%Unknown
70%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 12:52 | No view available |
| Corporation Tax | 3-11 (9 pages) | 95% | Processed | 2026-05-11 12:52 | No view available |
| Financial Statement | 12 (1 page) | 85% | Processed | 2026-05-11 12:52 | No view available |
| Corporation Tax | 13-14 (2 pages) | 95% | Processed | 2026-05-11 12:52 | No view available |
| Financial Statement | 15 (1 page) | 90% | Processed | 2026-05-11 12:52 | No view available |
| Corporation Tax | 16 (1 page) | 95% | Processed | 2026-05-11 12:52 | No view available |
| Corporation Tax | 17 (1 page) | 95% | Processed | 2026-05-11 12:52 | No view available |
| Corporation Tax | 18 (1 page) | 95% | Processed | 2026-05-11 12:52 | No view available |
| Corporation Tax | 20 (1 page) | 95% | Processed | 2026-05-11 12:52 | No view available |
| Financial Statement | 21 (1 page) | 90% | Processed | 2026-05-11 12:52 | No view available |
| P32 Payroll | 23 (1 page) | 95% | Processed | 2026-05-11 12:52 | No view available |
| Corporation Tax | 24 (1 page) | 95% | Processed | 2026-05-11 12:52 | No view available |
| Corporation Tax | 25 (1 page) | 95% | Processed | 2026-05-11 12:52 | No view available |
| Financial Statement | 27-28 (2 pages) | 95% | Processed | 2026-05-11 12:52 | No view available |
{
"document_groups": [
{
"avg_confidence": 0.9363636363636363,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "corporation_tax",
"end_page": 11,
"pages": [
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are key indicators of a corporation tax document. Additionally, it includes an IRmark number, which is specific to tax submissions.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKONMF0PjjJBOHwOgGSljxd6XVw2\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 ending 5 April 2025, and mentions the Self Assessment Online Service, which are key indicators of a corporation tax document. Additionally, it includes an IRmark number, which is specific to tax submissions.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778503879, 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=1383, total_tokens=1605, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_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 HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are key indicators of a corporation tax document. Additionally, it includes an IRmark number, which is specific to tax submissions.",
"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 ending 5 April 2025, and mentions the Self Assessment Online Service, which are key indicators of a corporation tax document. Additionally, it includes an IRmark number, which is specific to tax submissions.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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, \u0027Client Payment Advice\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are key indicators of a corporation tax document, likely related to a tax return or computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOPde3RCIq9k0AFoQTW061ELMvy\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, \\\u0027Client Payment Advice\\\u0027, \\\u0027Payments on Account\\\u0027, and \\\u0027Class 4 National Insurance Charge\\\u0027. These are key indicators of a corporation tax document, likely related to a tax return or computation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503881, 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=1036, total_tokens=1262, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_pred"
},
"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, \u0027Client Payment Advice\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are key indicators of a corporation tax document, likely related to a tax return or computation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis\u0027, \u0027Client Payment Advice\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are key indicators of a corporation tax document, likely related to a tax return or computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"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 is a tax return notice from HM Revenue \u0026 Customs (HMRC) for the tax year 2024-2025. It includes references to Self Assessment, UTR (Unique Taxpayer Reference), tax return deadlines, and penalties for late filing, which are all 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-DeKOQz3RdLIzkxWFesDuGWcBRCLIN\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 is a tax return notice from HM Revenue \u0026 Customs (HMRC) for the tax year 2024-2025. It includes references to Self Assessment, UTR (Unique Taxpayer Reference), tax return deadlines, and penalties for late filing, which are all 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=1778503882, 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=1734, total_tokens=1960, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_detai"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document is a tax return notice from HM Revenue \u0026 Customs (HMRC) for the tax year 2024-2025. It includes references to Self Assessment, UTR (Unique Taxpayer Reference), tax return deadlines, and penalties for late filing, which are all key indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document is a tax return notice from HM Revenue \u0026 Customs (HMRC) for the tax year 2024-2025. It includes references to Self Assessment, UTR (Unique Taxpayer Reference), tax return deadlines, and penalties for late filing, which are all 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 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, specifically a corporation tax return or related form.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOR28tKUKsW3am3s9F3mhrwDMzD\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 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, specifically a corporation tax return or related form.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503883, 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=1726, total_tokens=1959, completion_tokens_details=CompletionTokensDetails(accepted_prediction_t"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and 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, specifically a corporation tax return or related form.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and 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, specifically a corporation tax return or related form.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 4,
"text_length": 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, and \u0027Taxable Incapacity Benefit\u0027. Additionally, it includes references to specific tax boxes and forms like \u0027SA 100\u0027, which is associated with self-assessment tax returns in the UK. These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOTCcd5nqS2MKrzruYKrbk3XoEh\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, and \\\u0027Taxable Incapacity Benefit\\\u0027. Additionally, it includes references to specific tax boxes and forms like \\\u0027SA 100\\\u0027, which is associated with self-assessment tax returns in the UK. These are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503885, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=247, prompt_tokens=1645, total_tokens=1892, completion_tokens_details=Com"
},
"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, and \u0027Taxable Incapacity Benefit\u0027. Additionally, it includes references to specific tax boxes and forms like \u0027SA 100\u0027, which is associated with self-assessment tax returns in the UK. These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027State Pension\u0027, and \u0027Taxable Incapacity Benefit\u0027. Additionally, it includes references to specific tax boxes and forms like \u0027SA 100\u0027, which is associated with self-assessment tax returns in the UK. These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.15,
"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": 0.95,
"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 reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number (4639976719) and the mention of \u0027SA 100\u0027 (a UK self-assessment tax return form) strongly indicate that this is 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-DeKOUVmsXagke29quD1WUlodjgiM8\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 reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number (4639976719) and the mention of \\\u0027SA 100\\\u0027 (a UK self-assessment tax return form) strongly indicate that this is 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=1778503886, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=234, prompt_tokens=1604, total_tokens=1838, 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 tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number (4639976719) and the mention of \u0027SA 100\u0027 (a UK self-assessment tax return form) strongly indicate that this is a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number (4639976719) and the mention of \u0027SA 100\u0027 (a UK self-assessment tax return form) strongly indicate that this is a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"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 return document. The presence of \u0027SA 100\u0027 further confirms this 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-DeKOWGI5XtVyZ0UOMenx7InwLC4C5\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 return document. The presence of \\\u0027SA 100\\\u0027 further confirms this 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=1778503888, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=Comple"
},
"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 return document. The presence of \u0027SA 100\u0027 further confirms this 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 return document. The presence of \u0027SA 100\u0027 further confirms this 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 specific tax-related terms such as \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary.\u0027 These are strong indicators that the document is related to corporation tax filings or tax-related documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOXRtfQHMABJBMgCD1pjOTtITlA\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 specific tax-related terms such as \\\u0027Self Assessment Statement,\\\u0027 \\\u0027Tax Reference,\\\u0027 and \\\u0027Tax calculation summary.\\\u0027 These are strong indicators that the document is related to corporation tax filings or tax-related documentation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503889, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=217, prompt_tokens=1680, total_tokens=1897, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_to"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax returns, tax calculations, tax codes, and specific tax-related terms such as \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary.\u0027 These are strong indicators that the document is related to corporation tax filings or tax-related documentation.",
"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 specific tax-related terms such as \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary.\u0027 These are strong indicators that the document is related to corporation tax filings or tax-related documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "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": 0.95,
"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, \u0027Your tax adviser\u0027, \u0027SA100\u0027, and \u0027Client Ref\u0027. These are strong indicators of a tax-related document, specifically a corporation tax document. Additionally, the mention of \u0027Cooper Healey Coates Ltd\u0027 and the structured format aligns with typical corporation tax forms or 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-DeKOZvdH0pqXUw39tzAcdXFHrkmvv\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, \\\u0027Your tax adviser\\\u0027, \\\u0027SA100\\\u0027, and \\\u0027Client Ref\\\u0027. These are strong indicators of a tax-related document, specifically a corporation tax document. Additionally, the mention of \\\u0027Cooper Healey Coates Ltd\\\u0027 and the structured format aligns with typical corporation tax forms or 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=1778503891, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=235, prompt_tokens=1181, total_tokens=1416, completion_tokens_details=CompletionTokensDetails(accepted_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 tax-related information such as \u0027Tax Reference\u0027, \u0027Your tax adviser\u0027, \u0027SA100\u0027, and \u0027Client Ref\u0027. These are strong indicators of a tax-related document, specifically a corporation tax document. Additionally, the mention of \u0027Cooper Healey Coates Ltd\u0027 and the structured format aligns with typical corporation tax forms or related documents.",
"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, \u0027Your tax adviser\u0027, \u0027SA100\u0027, and \u0027Client Ref\u0027. These are strong indicators of a tax-related document, specifically a corporation tax document. Additionally, the mention of \u0027Cooper Healey Coates Ltd\u0027 and the structured format aligns with typical corporation tax forms or related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 9,
"text_length": 577
},
{
"confidence_score": 0.85,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. 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-DeKOavtbsRP0wtIY5uSLApGxKw7mg\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. 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=1778503892, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=197, prompt_tokens=1269, total_tokens=1466, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(audio_tokens=0, cached_tokens=0), latency_checkpoint={\u0027engine_tbt_ms\u0027: 4, \u0027engine_tt"
},
"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 key indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. These are key indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.0,
"total_score": 0.045
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 10,
"text_length": 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 business income reporting for tax purposes, aligning 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-DeKObq01QsF24OqvJfBSVxttvX1ZA\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 business income reporting for tax purposes, aligning 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=1778503893, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=217, prompt_tokens=1579, total_tokens=1796, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_token"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax 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 business income reporting for tax purposes, aligning with corporation tax documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and business income reporting for tax purposes, aligning with corporation tax documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.21741176470588236,
"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": 0.85,
"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": 12,
"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 breakdowns of business expenses, including categories such as wages, rent, repairs, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, balances, or account numbers that would classify it as a bank statement, credit card statement, or loan/mortgage statement. Additionally, it does not contain identifiers typical of invoices, payroll, or property-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-DeKOd1vKZPA4zLcUeZ8ybtU7zsThy\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 breakdowns of business expenses, including categories such as wages, rent, repairs, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, balances, or account numbers that would classify it as a bank statement, credit card statement, or loan/mortgage statement. Additionally, it does not contain identifiers typical of invoices, payroll, or property-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=1778503895, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=Completio"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"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 breakdowns of business expenses, including categories such as wages, rent, repairs, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, balances, or account numbers that would classify it as a bank statement, credit card statement, or loan/mortgage statement. Additionally, it does not contain identifiers typical of invoices, payroll, or property-related documents.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains detailed breakdowns of business expenses, including categories such as wages, rent, repairs, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, balances, or account numbers that would classify it as a bank statement, credit card statement, or loan/mortgage statement. Additionally, it does not contain identifiers typical of invoices, payroll, or property-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": "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
}
],
"start_page": 12
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "corporation_tax",
"end_page": 14,
"pages": [
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., box 47, box 48, box 49). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return document. These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOfqgVfbnZEpchSOOimQpsu45qs\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., box 47, box 48, box 49). Additionally, it includes a tax reference number and mentions \\\u0027Self-employment (full) notes,\\\u0027 which are indicative of a tax return document. These are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503897, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=247, prompt_tokens=1725, total_tokens=1972, completion_tokens_details=CompletionTokensDetails(ac"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., box 47, box 48, box 49). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return document. These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., box 47, box 48, box 49). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return document. These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 13,
"text_length": 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 profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOgOy4QzrcBs0ABufo4cfVdsmuh\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \\\u0027Self-employment (full) notes,\\\u0027 \\\u0027taxable profits,\\\u0027 and \\\u0027losses,\\\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503898, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=253, prompt_tokens=1732, total_tokens=1985, co"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.08470588235294117,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.08470588235294117
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 14,
"text_length": 2603
}
],
"start_page": 13
},
{
"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": 15,
"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 a balance sheet with detailed information about assets, liabilities, and capital of the business. Specific terms such as \u0027assets,\u0027 \u0027liabilities,\u0027 \u0027capital account,\u0027 and \u0027balance sheet\u0027 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-DeKOifBOpiGFexbiatFqmMwV54jRT\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 a balance sheet with detailed information about assets, liabilities, and capital of the business. Specific terms such as \\\u0027assets,\\\u0027 \\\u0027liabilities,\\\u0027 \\\u0027capital account,\\\u0027 and \\\u0027balance sheet\\\u0027 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=1778503900, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=211, prompt_tokens=1550, total_tokens=1761, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a balance sheet with detailed information about assets, liabilities, and capital of the business. Specific terms such as \u0027assets,\u0027 \u0027liabilities,\u0027 \u0027capital account,\u0027 and \u0027balance sheet\u0027 are key indicators of a financial statement.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital of the business. Specific terms such as \u0027assets,\u0027 \u0027liabilities,\u0027 \u0027capital account,\u0027 and \u0027balance sheet\u0027 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": 0.2374736842105263,
"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 specific tax-related instructions for self-employment and partnerships. 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-DeKOjHZ7ckinquqr0LEls61dDk3fj\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 specific tax-related instructions for self-employment and partnerships. 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=1778503901, 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=1236, total_tokens=1444, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDet"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and specific tax-related instructions for self-employment and partnerships. These are key 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-related information, including a tax reference number, National Insurance contributions (NICs), and specific tax-related instructions for self-employment and partnerships. These are key indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"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 calculation summary, Self Assessment, Unique Taxpayer Reference (UTR), and specific tax-related details such as Class 2 NICs, Class 4 NICs, Student Loan repayment, and Capital Gains Tax. These are strong indicators of a corporation tax-related document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOkc8SAPfy8ZYi1Frmfb0jNF67a\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax calculation summary, Self Assessment, Unique Taxpayer Reference (UTR), and specific tax-related details such as Class 2 NICs, Class 4 NICs, Student Loan repayment, and Capital Gains Tax. These are strong indicators of a corporation tax-related document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503902, 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_prediction_tokens=0), prom"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax calculation summary, Self Assessment, Unique Taxpayer Reference (UTR), and specific tax-related details such as Class 2 NICs, Class 4 NICs, Student Loan repayment, and Capital Gains Tax. These are strong indicators of a corporation tax-related document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax calculation summary, Self Assessment, Unique Taxpayer Reference (UTR), and specific tax-related details such as Class 2 NICs, Class 4 NICs, Student Loan repayment, and Capital Gains Tax. These are strong indicators of a corporation tax-related document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "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-DeKOmUOmlJx1z9YRUkvkmS5VXzBAQ\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=1778503904, 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 tax return schedule, computation summary, income, allowances, tax rates, and National Insurance Contributions (NIC). It also includes a tax reference number and mentions \u0027Tax Return Schedules\u0027 and \u0027Year of Assessment,\u0027 which are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOpYV5plPifqN63Ycwb1OKMUQ58\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 tax return schedule, computation summary, income, allowances, tax rates, and National Insurance Contributions (NIC). It also includes a tax reference number and mentions \\\u0027Tax Return Schedules\\\u0027 and \\\u0027Year of Assessment,\\\u0027 which are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503907, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=225, prompt_tokens=1293, total_tokens=1518, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_toke"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains tax-related information, including a tax return schedule, computation summary, income, allowances, tax rates, and National Insurance Contributions (NIC). It also includes a tax reference number and mentions \u0027Tax Return Schedules\u0027 and \u0027Year of Assessment,\u0027 which are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains tax-related information, including a tax return schedule, computation summary, income, allowances, tax rates, and National Insurance Contributions (NIC). It also includes a tax reference number and mentions \u0027Tax Return Schedules\u0027 and \u0027Year of Assessment,\u0027 which are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.1392857142857143,
"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": 0.9,
"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": 21,
"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, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOqvOZFRwwkXHarooBl5TmK6DRn\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, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503908, 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), prom"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 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": 23,
"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 payments required, payment dates, instalments, and amounts due, which are key indicators of an HMRC P32 Employer Payment Record. The structure and terminology align with tax obligations and employer payment records.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOtxI3GEWfieZPJDn4UeTcGInGt\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 payments required, payment dates, instalments, and amounts due, which are key indicators of an HMRC P32 Employer Payment Record. The structure and terminology align with tax obligations and employer payment records.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503911, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=210, prompt_tokens=1146, total_tokens=1356, 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 references to payments required, payment dates, instalments, and amounts due, which are key indicators of an HMRC P32 Employer Payment Record. The structure and terminology align with tax obligations and employer payment records.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to payments required, payment dates, instalments, and amounts due, which are key indicators of an HMRC P32 Employer Payment Record. The structure and terminology align with tax obligations and employer payment records.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "p32_payroll"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 23,
"text_length": 412
}
],
"start_page": 23
},
{
"avg_confidence": 0.925,
"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": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 \u0027Assessable Amount,\u0027 and \u0027Payment on Account Instalment.\u0027 These terms are indicative of a tax-related document, specifically related to corporation tax or self-assessment tax computations. The presence of a year-end date (\u0027Year ended 5 April 2025\u0027) and references to tax payments for the next year further support 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-DeKOvQi9ukMxG2GwUqqk9AvVoEeB6\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related terms such as \\\u0027Tax Due After Notional Taxes Paid,\\\u0027 \\\u0027Class 4 NIC Due,\\\u0027 \\\u0027Assessable Amount,\\\u0027 and \\\u0027Payment on Account Instalment.\\\u0027 These terms are indicative of a tax-related document, specifically related to corporation tax or self-assessment tax computations. The presence of a year-end date (\\\u0027Year ended 5 April 2025\\\u0027) and references to tax payments for the next year further support 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=1778503913, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=256, prompt_tokens=1130, total_t"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 \u0027Assessable Amount,\u0027 and \u0027Payment on Account Instalment.\u0027 These terms are indicative of a tax-related document, specifically related to corporation tax or self-assessment tax computations. The presence of a year-end date (\u0027Year ended 5 April 2025\u0027) and references to tax payments for the next year further support this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 \u0027Assessable Amount,\u0027 and \u0027Payment on Account Instalment.\u0027 These terms are indicative of a tax-related document, specifically related to corporation tax or self-assessment tax computations. The presence of a year-end date (\u0027Year ended 5 April 2025\u0027) and references to tax payments for the next year further support this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"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 strongly suggests it is related to corporation tax or self-assessment tax computations.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKOwmGb12fc0ozPR0i8faI9vSULz\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 strongly suggests it is related to corporation tax or self-assessment tax computations.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778503914, 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=1122, total_tokens=1355, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio"
},
"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 strongly suggests it is related to corporation tax or self-assessment tax computations.",
"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 strongly suggests it is related to corporation tax or self-assessment tax computations.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 24
},
{
"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 details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of other document types such as invoices, loan statements, or payroll summaries.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKOzVje5ilYISqCMfLOnbc7JltGD\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of other document types such as invoices, loan statements, or payroll summaries.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778503917, 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=1373, total_tokens=1593, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_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 details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of other document types such as invoices, loan statements, or payroll summaries.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of other document types such as invoices, loan statements, or payroll summaries.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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 assets, liabilities, and net business assets, as well as a detailed breakdown of net profit, adjustments, and capital account. These 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-DeKP0Ev7XdMFywXZQaBJMhVYEfEa2\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 assets, liabilities, and net business assets, as well as a detailed breakdown of net profit, adjustments, and capital account. These 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=1778503918, 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=1337, total_tokens=1541, 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="
},
"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 assets, liabilities, and net business assets, as well as a detailed breakdown of net profit, adjustments, and capital account. These 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 assets, liabilities, and net business assets, as well as a detailed breakdown of net profit, adjustments, and capital account. These 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 contains references to HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are key indicators of a corporation tax document. Additionally, it includes an IRmark number, which is specific to tax submissions.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKONMF0PjjJBOHwOgGSljxd6XVw2\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 ending 5 April 2025, and mentions the Self Assessment Online Service, which are key indicators of a corporation tax document. Additionally, it includes an IRmark number, which is specific to tax submissions.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778503879, 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=1383, total_tokens=1605, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_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 HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are key indicators of a corporation tax document. Additionally, it includes an IRmark number, which is specific to tax submissions.",
"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 ending 5 April 2025, and mentions the Self Assessment Online Service, which are key indicators of a corporation tax document. Additionally, it includes an IRmark number, which is specific to tax submissions.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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, \u0027Client Payment Advice\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are key indicators of a corporation tax document, likely related to a tax return or computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOPde3RCIq9k0AFoQTW061ELMvy\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, \\\u0027Client Payment Advice\\\u0027, \\\u0027Payments on Account\\\u0027, and \\\u0027Class 4 National Insurance Charge\\\u0027. These are key indicators of a corporation tax document, likely related to a tax return or computation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503881, 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=1036, total_tokens=1262, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_pred"
},
"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, \u0027Client Payment Advice\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are key indicators of a corporation tax document, likely related to a tax return or computation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis\u0027, \u0027Client Payment Advice\u0027, \u0027Payments on Account\u0027, and \u0027Class 4 National Insurance Charge\u0027. These are key indicators of a corporation tax document, likely related to a tax return or computation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"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 is a tax return notice from HM Revenue \u0026 Customs (HMRC) for the tax year 2024-2025. It includes references to Self Assessment, UTR (Unique Taxpayer Reference), tax return deadlines, and penalties for late filing, which are all 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-DeKOQz3RdLIzkxWFesDuGWcBRCLIN\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 is a tax return notice from HM Revenue \u0026 Customs (HMRC) for the tax year 2024-2025. It includes references to Self Assessment, UTR (Unique Taxpayer Reference), tax return deadlines, and penalties for late filing, which are all 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=1778503882, 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=1734, total_tokens=1960, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_detai"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document is a tax return notice from HM Revenue \u0026 Customs (HMRC) for the tax year 2024-2025. It includes references to Self Assessment, UTR (Unique Taxpayer Reference), tax return deadlines, and penalties for late filing, which are all key indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document is a tax return notice from HM Revenue \u0026 Customs (HMRC) for the tax year 2024-2025. It includes references to Self Assessment, UTR (Unique Taxpayer Reference), tax return deadlines, and penalties for late filing, which are all 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 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, specifically a corporation tax return or related form.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOR28tKUKsW3am3s9F3mhrwDMzD\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 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, specifically a corporation tax return or related form.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503883, 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=1726, total_tokens=1959, completion_tokens_details=CompletionTokensDetails(accepted_prediction_t"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and 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, specifically a corporation tax return or related form.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to a tax return, tax reference number, and various sections related to taxable income, capital gains, self-employment, partnerships, and 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, specifically a corporation tax return or related form.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 4,
"text_length": 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, and \u0027Taxable Incapacity Benefit\u0027. Additionally, it includes references to specific tax boxes and forms like \u0027SA 100\u0027, which is associated with self-assessment tax returns in the UK. These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOTCcd5nqS2MKrzruYKrbk3XoEh\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, and \\\u0027Taxable Incapacity Benefit\\\u0027. Additionally, it includes references to specific tax boxes and forms like \\\u0027SA 100\\\u0027, which is associated with self-assessment tax returns in the UK. These are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503885, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=247, prompt_tokens=1645, total_tokens=1892, completion_tokens_details=Com"
},
"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, and \u0027Taxable Incapacity Benefit\u0027. Additionally, it includes references to specific tax boxes and forms like \u0027SA 100\u0027, which is associated with self-assessment tax returns in the UK. These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027State Pension\u0027, and \u0027Taxable Incapacity Benefit\u0027. Additionally, it includes references to specific tax boxes and forms like \u0027SA 100\u0027, which is associated with self-assessment tax returns in the UK. These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.15,
"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": 0.95,
"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 reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number (4639976719) and the mention of \u0027SA 100\u0027 (a UK self-assessment tax return form) strongly indicate that this is 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-DeKOUVmsXagke29quD1WUlodjgiM8\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 reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number (4639976719) and the mention of \\\u0027SA 100\\\u0027 (a UK self-assessment tax return form) strongly indicate that this is 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=1778503886, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=234, prompt_tokens=1604, total_tokens=1838, 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 tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number (4639976719) and the mention of \u0027SA 100\u0027 (a UK self-assessment tax return form) strongly indicate that this is a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of a tax reference number (4639976719) and the mention of \u0027SA 100\u0027 (a UK self-assessment tax return form) strongly indicate that this is a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"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 return document. The presence of \u0027SA 100\u0027 further confirms this 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-DeKOWGI5XtVyZ0UOMenx7InwLC4C5\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 return document. The presence of \\\u0027SA 100\\\u0027 further confirms this 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=1778503888, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=Comple"
},
"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 return document. The presence of \u0027SA 100\u0027 further confirms this 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 return document. The presence of \u0027SA 100\u0027 further confirms this 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 specific tax-related terms such as \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary.\u0027 These are strong indicators that the document is related to corporation tax filings or tax-related documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOXRtfQHMABJBMgCD1pjOTtITlA\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 specific tax-related terms such as \\\u0027Self Assessment Statement,\\\u0027 \\\u0027Tax Reference,\\\u0027 and \\\u0027Tax calculation summary.\\\u0027 These are strong indicators that the document is related to corporation tax filings or tax-related documentation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503889, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=217, prompt_tokens=1680, total_tokens=1897, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_to"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax returns, tax calculations, tax codes, and specific tax-related terms such as \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary.\u0027 These are strong indicators that the document is related to corporation tax filings or tax-related documentation.",
"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 specific tax-related terms such as \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027Tax calculation summary.\u0027 These are strong indicators that the document is related to corporation tax filings or tax-related documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "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": 0.95,
"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, \u0027Your tax adviser\u0027, \u0027SA100\u0027, and \u0027Client Ref\u0027. These are strong indicators of a tax-related document, specifically a corporation tax document. Additionally, the mention of \u0027Cooper Healey Coates Ltd\u0027 and the structured format aligns with typical corporation tax forms or 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-DeKOZvdH0pqXUw39tzAcdXFHrkmvv\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, \\\u0027Your tax adviser\\\u0027, \\\u0027SA100\\\u0027, and \\\u0027Client Ref\\\u0027. These are strong indicators of a tax-related document, specifically a corporation tax document. Additionally, the mention of \\\u0027Cooper Healey Coates Ltd\\\u0027 and the structured format aligns with typical corporation tax forms or 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=1778503891, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=235, prompt_tokens=1181, total_tokens=1416, completion_tokens_details=CompletionTokensDetails(accepted_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 tax-related information such as \u0027Tax Reference\u0027, \u0027Your tax adviser\u0027, \u0027SA100\u0027, and \u0027Client Ref\u0027. These are strong indicators of a tax-related document, specifically a corporation tax document. Additionally, the mention of \u0027Cooper Healey Coates Ltd\u0027 and the structured format aligns with typical corporation tax forms or related documents.",
"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, \u0027Your tax adviser\u0027, \u0027SA100\u0027, and \u0027Client Ref\u0027. These are strong indicators of a tax-related document, specifically a corporation tax document. Additionally, the mention of \u0027Cooper Healey Coates Ltd\u0027 and the structured format aligns with typical corporation tax forms or related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 9,
"text_length": 577
},
{
"confidence_score": 0.85,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. 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-DeKOavtbsRP0wtIY5uSLApGxKw7mg\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. 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=1778503892, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=197, prompt_tokens=1269, total_tokens=1466, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(audio_tokens=0, cached_tokens=0), latency_checkpoint={\u0027engine_tbt_ms\u0027: 4, \u0027engine_tt"
},
"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 key indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. These are key indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.0,
"total_score": 0.045
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 10,
"text_length": 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 business income reporting for tax purposes, aligning 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-DeKObq01QsF24OqvJfBSVxttvX1ZA\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 business income reporting for tax purposes, aligning 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=1778503893, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=217, prompt_tokens=1579, total_tokens=1796, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_token"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax 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 business income reporting for tax purposes, aligning with corporation tax documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and business income reporting for tax purposes, aligning with corporation tax documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.21741176470588236,
"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": 0.85,
"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 breakdowns of business expenses, including categories such as wages, rent, repairs, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, balances, or account numbers that would classify it as a bank statement, credit card statement, or loan/mortgage statement. Additionally, it does not contain identifiers typical of invoices, payroll, or property-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-DeKOd1vKZPA4zLcUeZ8ybtU7zsThy\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 breakdowns of business expenses, including categories such as wages, rent, repairs, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, balances, or account numbers that would classify it as a bank statement, credit card statement, or loan/mortgage statement. Additionally, it does not contain identifiers typical of invoices, payroll, or property-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=1778503895, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=Completio"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"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 breakdowns of business expenses, including categories such as wages, rent, repairs, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, balances, or account numbers that would classify it as a bank statement, credit card statement, or loan/mortgage statement. Additionally, it does not contain identifiers typical of invoices, payroll, or property-related documents.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains detailed breakdowns of business expenses, including categories such as wages, rent, repairs, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, balances, or account numbers that would classify it as a bank statement, credit card statement, or loan/mortgage statement. Additionally, it does not contain identifiers typical of invoices, payroll, or property-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": "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": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., box 47, box 48, box 49). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return document. These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOfqgVfbnZEpchSOOimQpsu45qs\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., box 47, box 48, box 49). Additionally, it includes a tax reference number and mentions \\\u0027Self-employment (full) notes,\\\u0027 which are indicative of a tax return document. These are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503897, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=247, prompt_tokens=1725, total_tokens=1972, completion_tokens_details=CompletionTokensDetails(ac"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., box 47, box 48, box 49). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return document. These are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., box 47, box 48, box 49). Additionally, it includes a tax reference number and mentions \u0027Self-employment (full) notes,\u0027 which are indicative of a tax return document. These are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 13,
"text_length": 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 profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOgOy4QzrcBs0ABufo4cfVdsmuh\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \\\u0027Self-employment (full) notes,\\\u0027 \\\u0027taxable profits,\\\u0027 and \\\u0027losses,\\\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503898, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=253, prompt_tokens=1732, total_tokens=1985, co"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 73, box 76). It also includes terms like \u0027Self-employment (full) notes,\u0027 \u0027taxable profits,\u0027 and \u0027losses,\u0027 which are indicative of a tax-related document. Additionally, the presence of a tax reference number and client reference further supports the classification as a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.08470588235294117,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.08470588235294117
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 14,
"text_length": 2603
},
{
"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 a balance sheet with detailed information about assets, liabilities, and capital of the business. Specific terms such as \u0027assets,\u0027 \u0027liabilities,\u0027 \u0027capital account,\u0027 and \u0027balance sheet\u0027 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-DeKOifBOpiGFexbiatFqmMwV54jRT\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 a balance sheet with detailed information about assets, liabilities, and capital of the business. Specific terms such as \\\u0027assets,\\\u0027 \\\u0027liabilities,\\\u0027 \\\u0027capital account,\\\u0027 and \\\u0027balance sheet\\\u0027 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=1778503900, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=211, prompt_tokens=1550, total_tokens=1761, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a balance sheet with detailed information about assets, liabilities, and capital of the business. Specific terms such as \u0027assets,\u0027 \u0027liabilities,\u0027 \u0027capital account,\u0027 and \u0027balance sheet\u0027 are key indicators of a financial statement.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital of the business. Specific terms such as \u0027assets,\u0027 \u0027liabilities,\u0027 \u0027capital account,\u0027 and \u0027balance sheet\u0027 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": 0.2374736842105263,
"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 specific tax-related instructions for self-employment and partnerships. 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-DeKOjHZ7ckinquqr0LEls61dDk3fj\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 specific tax-related instructions for self-employment and partnerships. 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=1778503901, 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=1236, total_tokens=1444, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDet"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information, including a tax reference number, National Insurance contributions (NICs), and specific tax-related instructions for self-employment and partnerships. These are key 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-related information, including a tax reference number, National Insurance contributions (NICs), and specific tax-related instructions for self-employment and partnerships. These are key indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"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 calculation summary, Self Assessment, Unique Taxpayer Reference (UTR), and specific tax-related details such as Class 2 NICs, Class 4 NICs, Student Loan repayment, and Capital Gains Tax. These are strong indicators of a corporation tax-related document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOkc8SAPfy8ZYi1Frmfb0jNF67a\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax calculation summary, Self Assessment, Unique Taxpayer Reference (UTR), and specific tax-related details such as Class 2 NICs, Class 4 NICs, Student Loan repayment, and Capital Gains Tax. These are strong indicators of a corporation tax-related document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503902, 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_prediction_tokens=0), prom"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax calculation summary, Self Assessment, Unique Taxpayer Reference (UTR), and specific tax-related details such as Class 2 NICs, Class 4 NICs, Student Loan repayment, and Capital Gains Tax. These are strong indicators of a corporation tax-related document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax calculation summary, Self Assessment, Unique Taxpayer Reference (UTR), and specific tax-related details such as Class 2 NICs, Class 4 NICs, Student Loan repayment, and Capital Gains Tax. These are strong indicators of a corporation tax-related document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "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-DeKOmUOmlJx1z9YRUkvkmS5VXzBAQ\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=1778503904, 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 \u0027Form SA100 Overflow Report\u0027 and references self-employment and business details, which are typically associated with financial reporting. However, it does not explicitly match the structure or content of other categories like invoices, payroll, or tax returns. The lack of specific identifiers for other categories leads to the classification as a financial statement with moderate confidence.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOnraixwePm1EUJyXy3b9Qz0WHD\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 \\\u0027Form SA100 Overflow Report\\\u0027 and references self-employment and business details, which are typically associated with financial reporting. However, it does not explicitly match the structure or content of other categories like invoices, payroll, or tax returns. The lack of specific identifiers for other categories leads to the classification as a financial statement with moderate confidence.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503905, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=232, prompt_tokens=1055, total_tokens=1287, completion_tokens_details=Comp"
},
"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 \u0027Form SA100 Overflow Report\u0027 and references self-employment and business details, which are typically associated with financial reporting. However, it does not explicitly match the structure or content of other categories like invoices, payroll, or tax returns. The lack of specific identifiers for other categories leads to the classification as a financial statement with moderate confidence.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.7,\n \"reasoning\": \"The document mentions \u0027Form SA100 Overflow Report\u0027 and references self-employment and business details, which are typically associated with financial reporting. However, it does not explicitly match the structure or content of other categories like invoices, payroll, or tax returns. The lack of specific identifiers for other categories leads to the classification as a financial statement with moderate confidence.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "uncertain",
"pattern_based": {
"confidence": 0.065,
"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 tax return schedule, computation summary, income, allowances, tax rates, and National Insurance Contributions (NIC). It also includes a tax reference number and mentions \u0027Tax Return Schedules\u0027 and \u0027Year of Assessment,\u0027 which are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOpYV5plPifqN63Ycwb1OKMUQ58\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 tax return schedule, computation summary, income, allowances, tax rates, and National Insurance Contributions (NIC). It also includes a tax reference number and mentions \\\u0027Tax Return Schedules\\\u0027 and \\\u0027Year of Assessment,\\\u0027 which are strong indicators of a corporation tax document.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503907, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=225, prompt_tokens=1293, total_tokens=1518, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_toke"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains tax-related information, including a tax return schedule, computation summary, income, allowances, tax rates, and National Insurance Contributions (NIC). It also includes a tax reference number and mentions \u0027Tax Return Schedules\u0027 and \u0027Year of Assessment,\u0027 which are strong indicators of a corporation tax document.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains tax-related information, including a tax return schedule, computation summary, income, allowances, tax rates, and National Insurance Contributions (NIC). It also includes a tax reference number and mentions \u0027Tax Return Schedules\u0027 and \u0027Year of Assessment,\u0027 which are strong indicators of a corporation tax document.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.1392857142857143,
"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": 0.9,
"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, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOqvOZFRwwkXHarooBl5TmK6DRn\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, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503908, 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), prom"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. There are no indicators of transactions, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 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 documents. However, it does not match the specific structure of corporation tax returns or other 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-DeKOsyhxq4wXHSOxAWejc2rc7hQjb\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 documents. However, it does not match the specific structure of corporation tax returns or other 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=1778503910, 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 documents. However, it does not match the specific structure of corporation tax returns or other 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 documents. However, it does not match the specific structure of corporation tax returns or other 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 payments required, payment dates, instalments, and amounts due, which are key indicators of an HMRC P32 Employer Payment Record. The structure and terminology align with tax obligations and employer payment records.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DeKOtxI3GEWfieZPJDn4UeTcGInGt\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 payments required, payment dates, instalments, and amounts due, which are key indicators of an HMRC P32 Employer Payment Record. The structure and terminology align with tax obligations and employer payment records.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1778503911, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=210, prompt_tokens=1146, total_tokens=1356, 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 references to payments required, payment dates, instalments, and amounts due, which are key indicators of an HMRC P32 Employer Payment Record. The structure and terminology align with tax obligations and employer payment records.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to payments required, payment dates, instalments, and amounts due, which are key indicators of an HMRC P32 Employer Payment Record. The structure and terminology align with tax obligations and employer payment records.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "p32_payroll"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 23,
"text_length": 412
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 \u0027Assessable Amount,\u0027 and \u0027Payment on Account Instalment.\u0027 These terms are indicative of a tax-related document, specifically related to corporation tax or self-assessment tax computations. The presence of a year-end date (\u0027Year ended 5 April 2025\u0027) and references to tax payments for the next year further support 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-DeKOvQi9ukMxG2GwUqqk9AvVoEeB6\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related terms such as \\\u0027Tax Due After Notional Taxes Paid,\\\u0027 \\\u0027Class 4 NIC Due,\\\u0027 \\\u0027Assessable Amount,\\\u0027 and \\\u0027Payment on Account Instalment.\\\u0027 These terms are indicative of a tax-related document, specifically related to corporation tax or self-assessment tax computations. The presence of a year-end date (\\\u0027Year ended 5 April 2025\\\u0027) and references to tax payments for the next year further support 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=1778503913, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=256, prompt_tokens=1130, total_t"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 \u0027Assessable Amount,\u0027 and \u0027Payment on Account Instalment.\u0027 These terms are indicative of a tax-related document, specifically related to corporation tax or self-assessment tax computations. The presence of a year-end date (\u0027Year ended 5 April 2025\u0027) and references to tax payments for the next year further support this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 \u0027Assessable Amount,\u0027 and \u0027Payment on Account Instalment.\u0027 These terms are indicative of a tax-related document, specifically related to corporation tax or self-assessment tax computations. The presence of a year-end date (\u0027Year ended 5 April 2025\u0027) and references to tax payments for the next year further support this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"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 strongly suggests it is related to corporation tax or self-assessment tax computations.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKOwmGb12fc0ozPR0i8faI9vSULz\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 strongly suggests it is related to corporation tax or self-assessment tax computations.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778503914, 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=1122, total_tokens=1355, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio"
},
"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 strongly suggests it is related to corporation tax or self-assessment tax computations.",
"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 strongly suggests it is related to corporation tax or self-assessment tax computations.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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": 0.7,
"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 \u0027Year ended 5 April 2025\u0027 and \u0027Personal Allowances,\u0027 which are indicative of a financial summary or statement. However, it lacks specific indicators of other categories such as transactions, invoices, or loan details. The absence of detailed financial accounts or tax-specific terms reduces confidence slightly.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKOyA66ALEEFNiLgdvi606uN0h7h\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 \\\u0027Year ended 5 April 2025\\\u0027 and \\\u0027Personal Allowances,\\\u0027 which are indicative of a financial summary or statement. However, it lacks specific indicators of other categories such as transactions, invoices, or loan details. The absence of detailed financial accounts or tax-specific terms reduces confidence slightly.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778503916, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=224, prompt_tokens=1056, total_tokens=1280, 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 mentions \u0027Year ended 5 April 2025\u0027 and \u0027Personal Allowances,\u0027 which are indicative of a financial summary or statement. However, it lacks specific indicators of other categories such as transactions, invoices, or loan details. The absence of detailed financial accounts or tax-specific terms reduces confidence slightly.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.7,\n \"reasoning\": \"The document mentions \u0027Year ended 5 April 2025\u0027 and \u0027Personal Allowances,\u0027 which are indicative of a financial summary or statement. However, it lacks specific indicators of other categories such as transactions, invoices, or loan details. The absence of detailed financial accounts or tax-specific terms reduces confidence slightly.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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 details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of other document types such as invoices, loan statements, or payroll summaries.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DeKOzVje5ilYISqCMfLOnbc7JltGD\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of other document types such as invoices, loan statements, or payroll summaries.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1778503917, 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=1373, total_tokens=1593, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_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 details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of other document types such as invoices, loan statements, or payroll summaries.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and details about business income, expenses, and net profit. These are key indicators of a financial statement, specifically an income statement. There are no indicators of other document types such as invoices, loan statements, or payroll summaries.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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 assets, liabilities, and net business assets, as well as a detailed breakdown of net profit, adjustments, and capital account. These 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-DeKP0Ev7XdMFywXZQaBJMhVYEfEa2\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 assets, liabilities, and net business assets, as well as a detailed breakdown of net profit, adjustments, and capital account. These 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=1778503918, 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=1337, total_tokens=1541, 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="
},
"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 assets, liabilities, and net business assets, as well as a detailed breakdown of net profit, adjustments, and capital account. These 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 assets, liabilities, and net business assets, as well as a detailed breakdown of net profit, adjustments, and capital account. These 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": "2026-05-11T12:51:59.979420"
}