| Original Filename: | WIN002_Winter_D_M_SA100_24-25_1777294387.pdf |
| Upload Date: | April 27, 2026 at 12:58 PM |
| Total Pages: | 28 |
| Processing Status: | Completed |
| Total Documents: | 12 |
| Corporation Taxs: | 7 |
| Financial Statements: | 4 |
| P32 Payrolls: | 1 |
Corporation Tax
95%Corporation Tax
90%Corporation Tax
95%Corporation Tax
95%Corporation Tax
95%Corporation Tax
90%Corporation Tax
95%Corporation Tax
90%Corporation Tax
90%Corporation Tax
90%Corporation Tax
95%Financial Statement
90%Corporation Tax
95%Corporation Tax
95%Financial Statement
95%Corporation Tax
90%Corporation Tax
95%Corporation Tax
90%Corporation Tax
95%Financial Statement
85%Financial Statement
85%P32 Payroll
90%Corporation Tax
85%Corporation Tax
85%Unknown
70%Financial Statement
95%Financial Statement
95%Unknown
70%| Document Type | Pages | Confidence | Processing Status | Created | Actions |
|---|---|---|---|---|---|
| Corporation Tax | 1-2 (2 pages) | 95% | Processed | 2026-04-27 12:58 | No view available |
| Corporation Tax | 3-11 (9 pages) | 95% | Processed | 2026-04-27 12:58 | No view available |
| Financial Statement | 12 (1 page) | 90% | Processed | 2026-04-27 12:58 | No view available |
| Corporation Tax | 13-14 (2 pages) | 95% | Processed | 2026-04-27 12:58 | No view available |
| Financial Statement | 15 (1 page) | 95% | Processed | 2026-04-27 12:58 | No view available |
| Corporation Tax | 16 (1 page) | 95% | Processed | 2026-04-27 12:58 | No view available |
| Corporation Tax | 17-18 (2 pages) | 95% | Processed | 2026-04-27 12:58 | No view available |
| Corporation Tax | 19 (1 page) | 90% | Processed | 2026-04-27 12:58 | No view available |
| Financial Statement | 20-21 (2 pages) | 85% | Processed | 2026-04-27 12:58 | No view available |
| P32 Payroll | 22 (1 page) | 90% | Processed | 2026-04-27 12:58 | No view available |
| Corporation Tax | 23-24 (2 pages) | 85% | Processed | 2026-04-27 12:58 | No view available |
| Financial Statement | 26-27 (2 pages) | 95% | Processed | 2026-04-27 12:58 | No view available |
{
"document_groups": [
{
"avg_confidence": 0.9272727272727272,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "corporation_tax",
"end_page": 11,
"pages": [
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, the presence of an IRmark number and references to tax repayment claims 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-DZFoU7jrMtDzzez5J7yJQSHsnHPF8\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, the presence of an IRmark number and references to tax repayment claims 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=1777294638, 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=1385, total_tokens=1610, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rej"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, the presence of an IRmark number and references to tax repayment claims further support this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, the presence of an IRmark number and references to tax repayment claims 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": "both_agree",
"pattern_based": {
"confidence": 0.13270588235294117,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.1,
"total_score": 0.13270588235294117
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "corporation_tax"
}
},
"page_number": 1,
"text_length": 1900
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis,\u0027 \u0027Payments on Account,\u0027 and \u0027Class 4 National Insurance Charge,\u0027 which are indicative of a corporation tax return or related tax documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoWQKBViE9a1qW1hmwPWYwCdn3Z\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document mentions \\\u00272024/25 Tax Return\\\u0027 and includes references to tax-related schedules such as \\\u0027Rate Band Analysis,\\\u0027 \\\u0027Payments on Account,\\\u0027 and \\\u0027Class 4 National Insurance Charge,\\\u0027 which are indicative of a corporation tax return or related tax documentation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294640, 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=1040, total_tokens=1255, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptToken"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis,\u0027 \u0027Payments on Account,\u0027 and \u0027Class 4 National Insurance Charge,\u0027 which are indicative of a corporation tax return or related tax documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis,\u0027 \u0027Payments on Account,\u0027 and \u0027Class 4 National Insurance Charge,\u0027 which are indicative of a corporation tax return or related tax documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 2,
"text_length": 213
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax return, self-assessment, UTR (Unique Taxpayer Reference), and tax year details, which are key indicators of a corporation tax document. Additionally, it includes instructions for filing a tax return and penalties for late submission, which are specific to tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoYyoIxuABT11hM8VHt7giH8Zeg\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax return, self-assessment, UTR (Unique Taxpayer Reference), and tax year details, which are key indicators of a corporation tax document. Additionally, it includes instructions for filing a tax return and penalties for late submission, which are specific to tax-related documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294642, 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=1737, total_tokens=1963, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_p"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax return, self-assessment, UTR (Unique Taxpayer Reference), and tax year details, which are key indicators of a corporation tax document. Additionally, it includes instructions for filing a tax return and penalties for late submission, which are specific to tax-related documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax return, self-assessment, UTR (Unique Taxpayer Reference), and tax year details, which are key indicators of a corporation tax document. Additionally, it includes instructions for filing a tax return and penalties for late submission, which are specific to tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.1807058823529412,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.2,
"total_score": 0.1807058823529412
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 3,
"text_length": 2711
},
{
"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-DZFoZHCKnGE4SXD2tbA9vvxCe7RlL\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=1777294643, 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=1731, total_tokens=1963, completion_tokens_details=CompletionTokensDetails(accepted_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 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": 4065
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027State Pension\u0027, and \u0027Taxable Incapacity Benefit\u0027. Additionally, it includes references to specific tax boxes and forms (e.g., \u0027SA100 2025 Page 3 of 16\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-DZFobZBcbsuFboFn7x2VBjqKJ33kW\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 (e.g., \\\u0027SA100 2025 Page 3 of 16\\\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=1777294645, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=254, prompt_tokens=1631, total_tokens=1885, completion_tokens_details="
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \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 (e.g., \u0027SA100 2025 Page 3 of 16\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 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 (e.g., \u0027SA100 2025 Page 3 of 16\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": 0.15,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.125,
"total_score": 0.15
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.030303030303030304,
"total_score": 0.03681818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 5,
"text_length": 2225
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid payments\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoc0NLK8jGlbWDmIZgxC8BCsOV3\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \\\u0027Tax Reference\\\u0027, \\\u0027Gift Aid payments\\\u0027, and \\\u0027SA 100\\\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294646, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=228, prompt_tokens=1597, total_tokens=1825, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, "
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid payments\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid payments\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.030303030303030304,
"total_score": 0.03681818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 6,
"text_length": 2173
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFodp043xIAiUJzf4A4l2p2pARk0\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Student Loan and Postgraduate Loan repayments\\\u0027, \\\u0027High Income Child Benefit Charge\\\u0027, and \\\u0027Marriage Allowance\\\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \\\u0027SA 100\\\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294647, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usag"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11454545454545456,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.09090909090909091,
"total_score": 0.11454545454545456
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.030303030303030304,
"total_score": 0.06681818181818183
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "loan_statement"
}
},
"page_number": 7,
"text_length": 2182
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, and HMRC-related information, which are key indicators of a corporation tax document. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027SA 100\u0027 further confirm this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFog6Xlj93bgr2Tc2bX4xmXQ7k9u\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax returns, tax calculations, and HMRC-related information, which are key indicators of a corporation tax document. Specific mentions of \\\u0027Self Assessment Statement,\\\u0027 \\\u0027Tax Reference,\\\u0027 and \\\u0027SA 100\\\u0027 further confirm this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294650, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1676, total_tokens=1890, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptT"
},
"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, and HMRC-related information, which are key indicators of a corporation tax document. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027SA 100\u0027 further confirm this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, and HMRC-related information, which are key indicators of a corporation tax document. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027SA 100\u0027 further confirm this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.1392857142857143,
"features": {
"bank_statement": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.0,
"total_score": 0.10588235294117647
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.14285714285714285,
"total_score": 0.1392857142857143
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.030303030303030304,
"total_score": 0.05181818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 8,
"text_length": 2555
},
{
"confidence_score": 0.9,
"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, a tax adviser\u0027s name and address, and a client reference. Additionally, the mention of \u0027SA 100\u0027 suggests it is related to self-assessment tax returns, which aligns with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoh1yNq8ZWNJzXLHJrDI12WCz7F\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, a tax adviser\\\u0027s name and address, and a client reference. Additionally, the mention of \\\u0027SA 100\\\u0027 suggests it is related to self-assessment tax returns, which aligns with corporation tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294651, 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=1179, total_tokens=1399, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_toke"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information, including a tax reference number, a tax adviser\u0027s name and address, and a client reference. Additionally, the mention of \u0027SA 100\u0027 suggests it is related to self-assessment tax returns, which aligns with corporation tax documents.",
"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, a tax adviser\u0027s name and address, and a client reference. Additionally, the mention of \u0027SA 100\u0027 suggests it is related to self-assessment tax returns, which aligns with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 9,
"text_length": 577
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. These are 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-DZFoiwqBmYvZn96ROjebb1ZLM7b8p\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. These are 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=1777294652, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=198, prompt_tokens=1264, total_tokens=1462, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(audio_tokens=0, cached_tokens=1152)), prompt_filter_results=[{\u0027prompt_index\u0027: 0, \u0027co"
},
"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\": 0.9,\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": 1071
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and tax reporting, aligning with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFokQqWrd0VSERMceXECSxnVmCIX\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and tax reporting, aligning with corporation tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294654, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1566, total_tokens=1780, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(au"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and tax reporting, aligning with corporation tax documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and tax reporting, aligning with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.21741176470588236,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.23529411764705882,
"pattern_score": 0.1,
"total_score": 0.21741176470588236
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.15789473684210525,
"pattern_score": 0.0,
"total_score": 0.11368421052631578
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.0,
"total_score": 0.13
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 11,
"text_length": 1833
}
],
"start_page": 1
},
{
"avg_confidence": 0.9,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "financial_statement",
"end_page": 12,
"pages": [
{
"confidence_score": 0.9,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data related to business expenses, including categories such as wages, rent, advertising, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFol1oHLJahwTT0EirBC4hcejwcf\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains detailed financial data related to business expenses, including categories such as wages, rent, advertising, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294655, 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=1586, total_tokens=1812, completion_tokens_details=CompletionTokensDetails(accepted_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 contains detailed financial data related to business expenses, including categories such as wages, rent, advertising, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data related to business expenses, including categories such as wages, rent, advertising, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.03529411764705882
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.0,
"total_score": 0.04
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 12,
"text_length": 1495
}
],
"start_page": 12
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "corporation_tax",
"end_page": 14,
"pages": [
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., box 47, box 48, box 63). These are indicative of a corporation tax document, likely part of 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-DZFomxN3DeDgY6WRPH09l4VfcmUTA\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 63). These are indicative of a corporation tax document, likely part of 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=1777294656, 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=1753, total_tokens=1980, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_toke"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to 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 63). These are indicative of a corporation tax document, likely part of a tax return or computation.",
"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 63). These are indicative of a corporation tax document, likely part of 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": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 13,
"text_length": 2581
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 68, box 73.3). It also mentions \u0027Self-employment (full) notes\u0027 and \u0027SA103F,\u0027 which are associated with UK self-assessment tax forms. 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-DZFooVGqv45XzO1gTZyleWukS31vY\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 68, box 73.3). It also mentions \\\u0027Self-employment (full) notes\\\u0027 and \\\u0027SA103F,\\\u0027 which are associated with UK self-assessment tax forms. 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=1777294658, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=241, prompt_tokens=1735, total_tokens=1976, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, re"
},
"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 68, box 73.3). It also mentions \u0027Self-employment (full) notes\u0027 and \u0027SA103F,\u0027 which are associated with UK self-assessment tax forms. 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 taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 68, box 73.3). It also mentions \u0027Self-employment (full) notes\u0027 and \u0027SA103F,\u0027 which are associated with UK self-assessment tax forms. 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": 0.08470588235294117,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.08470588235294117
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 14,
"text_length": 2614
}
],
"start_page": 13
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "financial_statement",
"end_page": 15,
"pages": [
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial statements.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFopKOYa3KkHa0118jh5UpLEyZxH\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital accounts. Key indicators include terms like \\\u0027Balance sheet,\\\u0027 \\\u0027Assets,\\\u0027 \\\u0027Liabilities,\\\u0027 \\\u0027Net business assets,\\\u0027 and \\\u0027Capital account,\\\u0027 which are characteristic of financial statements.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294659, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1545, total_tokens=1759, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), promp"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a balance sheet with detailed information about assets, liabilities, and capital accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial statements.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial statements.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.2374736842105263,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.2631578947368421,
"pattern_score": 0.1,
"total_score": 0.2374736842105263
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 15,
"text_length": 1436
}
],
"start_page": 15
},
{
"avg_confidence": 0.925,
"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": 19,
"pages": [
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax-related information, such as \u0027Class 2 and Class 4 National Insurance contributions,\u0027 \u0027profits from self-employment,\u0027 and \u0027Tax Reference.\u0027 These are strong indicators of a tax-related document, specifically related to self-assessment or corporation tax. The presence of terms like \u0027SA103F\u0027 and \u0027Tax Reference\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZForwn5v0PsaAgTLIx74GeQl5cKR\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax-related information, such as \\\u0027Class 2 and Class 4 National Insurance contributions,\\\u0027 \\\u0027profits from self-employment,\\\u0027 and \\\u0027Tax Reference.\\\u0027 These are strong indicators of a tax-related document, specifically related to self-assessment or corporation tax. The presence of terms like \\\u0027SA103F\\\u0027 and \\\u0027Tax Reference\\\u0027 further supports this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294661, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=238, prompt_tokens=1211, total_tokens=1449, completion_tokens_details=CompletionTokensDetails(accept"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information, such as \u0027Class 2 and Class 4 National Insurance contributions,\u0027 \u0027profits from self-employment,\u0027 and \u0027Tax Reference.\u0027 These are strong indicators of a tax-related document, specifically related to self-assessment or corporation tax. The presence of terms like \u0027SA103F\u0027 and \u0027Tax Reference\u0027 further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax-related information, such as \u0027Class 2 and Class 4 National Insurance contributions,\u0027 \u0027profits from self-employment,\u0027 and \u0027Tax Reference.\u0027 These are strong indicators of a tax-related document, specifically related to self-assessment or corporation tax. The presence of terms like \u0027SA103F\u0027 and \u0027Tax Reference\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 16,
"text_length": 766
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax calculations, Self Assessment, and specific tax-related terms such as \u0027Tax calculation summary,\u0027 \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 \u0027Student Loan repayment,\u0027 and \u0027Postgraduate Loan repayment.\u0027 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-DZFostYDv5Frk1QGVqKKd9cNeKabb\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 calculations, Self Assessment, and specific tax-related terms such as \\\u0027Tax calculation summary,\\\u0027 \\\u0027Class 2 NICs,\\\u0027 \\\u0027Class 4 NICs,\\\u0027 \\\u0027Student Loan repayment,\\\u0027 and \\\u0027Postgraduate Loan repayment.\\\u0027 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=1777294662, 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=1682, total_tokens=1908, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tok"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax calculations, Self Assessment, and specific tax-related terms such as \u0027Tax calculation summary,\u0027 \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 \u0027Student Loan repayment,\u0027 and \u0027Postgraduate Loan repayment.\u0027 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 calculations, Self Assessment, and specific tax-related terms such as \u0027Tax calculation summary,\u0027 \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 \u0027Student Loan repayment,\u0027 and \u0027Postgraduate Loan repayment.\u0027 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": 2275
},
{
"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 \u0027Blind person\u0027s surplus allowance,\u0027 \u0027married couple\u0027s surplus allowance,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document or related tax filing.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFouazEtfT1MoeVHTfXTU3dn1xOU\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 \\\u0027Blind person\\\u0027s surplus allowance,\\\u0027 \\\u0027married couple\\\u0027s surplus allowance,\\\u0027 and \\\u0027self-assessment statement of account.\\\u0027 These are strong indicators of a corporation tax document or related tax filing.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294664, 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=1329, total_tokens=1563, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027Blind person\u0027s surplus allowance,\u0027 \u0027married couple\u0027s surplus allowance,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document or related tax filing.",
"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 \u0027Blind person\u0027s surplus allowance,\u0027 \u0027married couple\u0027s surplus allowance,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document or related tax filing.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.07058823529411765,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "bank_statement"
}
},
"page_number": 18,
"text_length": 1362
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains tax-related information, including a tax reference number, National Insurance (NI) number, income and outgoings, tax rates, and a computation summary for the year of assessment. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Income Tax,\u0027 \u0027NIC,\u0027 and \u0027Taxable Income\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFovCWVsU05E1mLlcTWOnSnTFxI7\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 reference number, National Insurance (NI) number, income and outgoings, tax rates, and a computation summary for the year of assessment. These are key indicators of a tax return or corporation tax document. The presence of terms like \\\u0027Tax Return Schedules,\\\u0027 \\\u0027Income Tax,\\\u0027 \\\u0027NIC,\\\u0027 and \\\u0027Taxable Income\\\u0027 further supports this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294665, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=244, prompt_tokens=1280, total_tokens=1524, completion_tokens_details=CompletionTokensD"
},
"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 reference number, National Insurance (NI) number, income and outgoings, tax rates, and a computation summary for the year of assessment. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Income Tax,\u0027 \u0027NIC,\u0027 and \u0027Taxable Income\u0027 further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains tax-related information, including a tax reference number, National Insurance (NI) number, income and outgoings, tax rates, and a computation summary for the year of assessment. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Income Tax,\u0027 \u0027NIC,\u0027 and \u0027Taxable Income\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.030303030303030304,
"total_score": 0.08181818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 19,
"text_length": 878
}
],
"start_page": 16
},
{
"avg_confidence": 0.85,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "financial_statement",
"end_page": 21,
"pages": [
{
"confidence_score": 0.85,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. It does not include transaction details, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoxXvV8sDn8x4imJKRPiBtZlBHh\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. It does not include transaction details, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294667, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1961, total_tokens=2175, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prom"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. It does not include transaction details, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. It does not include transaction details, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 20,
"text_length": 1708
},
{
"confidence_score": 0.85,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, investors\u0027 relief gains, and other financial metrics typically found in financial statements. There are no indicators of transactions, invoices, payroll, or property-related details, which rules out other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoyFAQHCCV8NqyBOflJMxaBlQQD\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, investors\\\u0027 relief gains, and other financial metrics typically found in financial statements. There are no indicators of transactions, invoices, payroll, or property-related details, which rules out other categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294668, 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=1117, total_tokens=1334, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_predic"
},
"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, investors\u0027 relief gains, and other financial metrics typically found in financial statements. There are no indicators of transactions, invoices, payroll, or property-related details, which rules out other categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, investors\u0027 relief gains, and other financial metrics typically found in financial statements. There are no indicators of transactions, invoices, payroll, or property-related details, which rules out other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 21,
"text_length": 402
}
],
"start_page": 20
},
{
"avg_confidence": 0.9,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "p32_payroll",
"end_page": 22,
"pages": [
{
"confidence_score": 0.9,
"document_type": "p32_payroll",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure and terminology align with tax obligations and payment schedules typically found in P32 payroll documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFp0XI51dk6JUYZFTfITLdsZlLdP\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"p32_payroll\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to \\\u0027Client Payment Advice,\\\u0027 \\\u0027Payments Required,\\\u0027 \\\u0027Payment Dates,\\\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure and terminology align with tax obligations and payment schedules typically found in P32 payroll documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294670, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=221, prompt_tokens=1137, total_tokens=1358, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_predic"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure and terminology align with tax obligations and payment schedules typically found in P32 payroll documents.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure and terminology align with tax obligations and payment schedules typically found in P32 payroll documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "p32_payroll"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 22,
"text_length": 400
}
],
"start_page": 22
},
{
"avg_confidence": 0.85,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "corporation_tax",
"end_page": 24,
"pages": [
{
"confidence_score": 0.85,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account.\u0027 These are indicative of a tax-related document, specifically related to individual or corporate tax obligations. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting, and the format resembles a tax computation or summary.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DZFp1zxm4O8zRRayzGTKtzkh42kT7\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document references tax-related terms such as \\\u0027Tax Due After Notional Taxes Paid,\\\u0027 \\\u0027Class 4 NIC Due,\\\u0027 and \\\u0027Payments on Account.\\\u0027 These are indicative of a tax-related document, specifically related to individual or corporate tax obligations. The mention of \\\u0027Year ended 5 April 2025\\\u0027 aligns with tax year reporting, and the format resembles a tax computation or summary.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1777294671, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=240, prompt_tokens=1111, total_tokens=1351, completion_tokens_details=CompletionTokensDetails(accepted_prediction_"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account.\u0027 These are indicative of a tax-related document, specifically related to individual or corporate tax obligations. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting, and the format resembles a tax computation or summary.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account.\u0027 These are indicative of a tax-related document, specifically related to individual or corporate tax obligations. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting, and the format resembles a tax computation or summary.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 23,
"text_length": 374
},
{
"confidence_score": 0.85,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document 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 \u0027Total Class 4 NIC charge\u0027 suggests it is related to tax obligations, specifically for self-employed individuals. While it does not explicitly mention \u0027CT600\u0027 or \u0027corporation tax return\u0027, the content aligns most closely with the \u0027corporation_tax\u0027 category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFp372UnBUyjNe78yMxTh8UrS4Rh\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 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 \\\u0027Total Class 4 NIC charge\\\u0027 suggests it is related to tax obligations, specifically for self-employed individuals. While it does not explicitly mention \\\u0027CT600\\\u0027 or \\\u0027corporation tax return\\\u0027, the content aligns most closely with the \\\u0027corporation_tax\\\u0027 category.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294673, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=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 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 \u0027Total Class 4 NIC charge\u0027 suggests it is related to tax obligations, specifically for self-employed individuals. While it does not explicitly mention \u0027CT600\u0027 or \u0027corporation tax return\u0027, the content aligns most closely with the \u0027corporation_tax\u0027 category.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\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 \u0027Total Class 4 NIC charge\u0027 suggests it is related to tax obligations, specifically for self-employed individuals. While it does not explicitly mention \u0027CT600\u0027 or \u0027corporation tax return\u0027, the content aligns most closely with the \u0027corporation_tax\u0027 category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 24,
"text_length": 359
}
],
"start_page": 23
},
{
"avg_confidence": 0.95,
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"document_type": "financial_statement",
"end_page": 27,
"pages": [
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and detailed financial data such as sales income, expenses, net profit, and capital allowances. These are key indicators of a financial statement, specifically an income statement for a sole proprietorship.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DZFp66w5fnDURwY1vpNT4snll7Grz\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and detailed financial data such as sales income, expenses, net profit, and capital allowances. These are key indicators of a financial statement, specifically an income statement for a sole proprietorship.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1777294676, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=212, prompt_tokens=1410, total_tokens=1622, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=Promp"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a profit and loss account, tax adjustments, and detailed financial data such as sales income, expenses, net profit, and capital allowances. These are key indicators of a financial statement, specifically an income statement for a sole proprietorship.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and detailed financial data such as sales income, expenses, net profit, and capital allowances. These are key indicators of a financial statement, specifically an income statement for a sole proprietorship.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.1616842105263158,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.0,
"total_score": 0.12705882352941175
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.15789473684210525,
"pattern_score": 0.1,
"total_score": 0.1616842105263158
},
"hp_agreement": {
"keyword_score": 0.038461538461538464,
"pattern_score": 0.0,
"total_score": 0.023076923076923078
},
"invoice": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.0,
"total_score": 0.09999999999999999
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "financial_statement"
}
},
"page_number": 26,
"text_length": 1304
},
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet summary with assets, liabilities, and net business assets, as well as a capital account and profit/loss adjustments. These are key indicators of a financial statement, specifically a balance sheet and income 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-DZFp8YcyvLGfrxPdRegBNREjsCqkS\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 capital account and profit/loss adjustments. These are key indicators of a financial statement, specifically a balance sheet and income 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=1777294678, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=209, prompt_tokens=1378, total_tokens=1587, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(a"
},
"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 capital account and profit/loss adjustments. These are key indicators of a financial statement, specifically a balance sheet and income 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 capital account and profit/loss adjustments. These are key indicators of a financial statement, specifically a balance sheet and income statement.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.25768421052631574,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.15789473684210525,
"pattern_score": 0.3,
"total_score": 0.25768421052631574
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "financial_statement"
}
},
"page_number": 27,
"text_length": 1111
}
],
"start_page": 26
}
],
"page_classifications": [
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, the presence of an IRmark number and references to tax repayment claims 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-DZFoU7jrMtDzzez5J7yJQSHsnHPF8\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, the presence of an IRmark number and references to tax repayment claims 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=1777294638, 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=1385, total_tokens=1610, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rej"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, the presence of an IRmark number and references to tax repayment claims further support this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document references HM Revenue \u0026 Customs, a tax return for the year ending 5 April 2025, and mentions the Self Assessment Online Service, which are strong indicators of a corporation tax-related document. Additionally, the presence of an IRmark number and references to tax repayment claims 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": "both_agree",
"pattern_based": {
"confidence": 0.13270588235294117,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.1,
"total_score": 0.13270588235294117
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "corporation_tax"
}
},
"page_number": 1,
"text_length": 1900
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis,\u0027 \u0027Payments on Account,\u0027 and \u0027Class 4 National Insurance Charge,\u0027 which are indicative of a corporation tax return or related tax documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoWQKBViE9a1qW1hmwPWYwCdn3Z\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document mentions \\\u00272024/25 Tax Return\\\u0027 and includes references to tax-related schedules such as \\\u0027Rate Band Analysis,\\\u0027 \\\u0027Payments on Account,\\\u0027 and \\\u0027Class 4 National Insurance Charge,\\\u0027 which are indicative of a corporation tax return or related tax documentation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294640, 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=1040, total_tokens=1255, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptToken"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis,\u0027 \u0027Payments on Account,\u0027 and \u0027Class 4 National Insurance Charge,\u0027 which are indicative of a corporation tax return or related tax documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document mentions \u00272024/25 Tax Return\u0027 and includes references to tax-related schedules such as \u0027Rate Band Analysis,\u0027 \u0027Payments on Account,\u0027 and \u0027Class 4 National Insurance Charge,\u0027 which are indicative of a corporation tax return or related tax documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 2,
"text_length": 213
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax return, self-assessment, UTR (Unique Taxpayer Reference), and tax year details, which are key indicators of a corporation tax document. Additionally, it includes instructions for filing a tax return and penalties for late submission, which are specific to tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoYyoIxuABT11hM8VHt7giH8Zeg\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax return, self-assessment, UTR (Unique Taxpayer Reference), and tax year details, which are key indicators of a corporation tax document. Additionally, it includes instructions for filing a tax return and penalties for late submission, which are specific to tax-related documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294642, 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=1737, total_tokens=1963, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_p"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax return, self-assessment, UTR (Unique Taxpayer Reference), and tax year details, which are key indicators of a corporation tax document. Additionally, it includes instructions for filing a tax return and penalties for late submission, which are specific to tax-related documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax return, self-assessment, UTR (Unique Taxpayer Reference), and tax year details, which are key indicators of a corporation tax document. Additionally, it includes instructions for filing a tax return and penalties for late submission, which are specific to tax-related documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.1807058823529412,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.2,
"total_score": 0.1807058823529412
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 3,
"text_length": 2711
},
{
"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-DZFoZHCKnGE4SXD2tbA9vvxCe7RlL\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=1777294643, 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=1731, total_tokens=1963, completion_tokens_details=CompletionTokensDetails(accepted_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 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": 4065
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Taxed UK interest\u0027, \u0027Untaxed UK interest\u0027, \u0027Foreign dividends\u0027, \u0027State Pension\u0027, and \u0027Taxable Incapacity Benefit\u0027. Additionally, it includes references to specific tax boxes and forms (e.g., \u0027SA100 2025 Page 3 of 16\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-DZFobZBcbsuFboFn7x2VBjqKJ33kW\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 (e.g., \\\u0027SA100 2025 Page 3 of 16\\\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=1777294645, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=254, prompt_tokens=1631, total_tokens=1885, completion_tokens_details="
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \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 (e.g., \u0027SA100 2025 Page 3 of 16\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 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 (e.g., \u0027SA100 2025 Page 3 of 16\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": 0.15,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.125,
"total_score": 0.15
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.030303030303030304,
"total_score": 0.03681818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 5,
"text_length": 2225
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid payments\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoc0NLK8jGlbWDmIZgxC8BCsOV3\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \\\u0027Tax Reference\\\u0027, \\\u0027Gift Aid payments\\\u0027, and \\\u0027SA 100\\\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294646, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=228, prompt_tokens=1597, total_tokens=1825, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, "
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid payments\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax reliefs, pension contributions, charitable giving, and allowances, which are typically associated with tax-related documents. Additionally, the presence of terms like \u0027Tax Reference\u0027, \u0027Gift Aid payments\u0027, and \u0027SA 100\u0027 strongly suggests this is related to a corporation tax return or similar tax documentation.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.030303030303030304,
"total_score": 0.03681818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 6,
"text_length": 2173
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFodp043xIAiUJzf4A4l2p2pARk0\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to tax-related information such as \\\u0027Tax Reference\\\u0027, \\\u0027Student Loan and Postgraduate Loan repayments\\\u0027, \\\u0027High Income Child Benefit Charge\\\u0027, and \\\u0027Marriage Allowance\\\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \\\u0027SA 100\\\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294647, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usag"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related information such as \u0027Tax Reference\u0027, \u0027Student Loan and Postgraduate Loan repayments\u0027, \u0027High Income Child Benefit Charge\u0027, and \u0027Marriage Allowance\u0027. Additionally, it includes specific tax-related details like income thresholds, child benefit amounts, and personal allowance transfers, which are indicative of a corporation tax or personal tax document. The presence of \u0027SA 100\u0027 further supports this classification as it is a self-assessment tax return form used in the UK.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11454545454545456,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.09090909090909091,
"total_score": 0.11454545454545456
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.041666666666666664,
"pattern_score": 0.030303030303030304,
"total_score": 0.06681818181818183
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "loan_statement"
}
},
"page_number": 7,
"text_length": 2182
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, and HMRC-related information, which are key indicators of a corporation tax document. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027SA 100\u0027 further confirm this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFog6Xlj93bgr2Tc2bX4xmXQ7k9u\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax returns, tax calculations, and HMRC-related information, which are key indicators of a corporation tax document. Specific mentions of \\\u0027Self Assessment Statement,\\\u0027 \\\u0027Tax Reference,\\\u0027 and \\\u0027SA 100\\\u0027 further confirm this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294650, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1676, total_tokens=1890, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptT"
},
"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, and HMRC-related information, which are key indicators of a corporation tax document. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027SA 100\u0027 further confirm this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax returns, tax calculations, and HMRC-related information, which are key indicators of a corporation tax document. Specific mentions of \u0027Self Assessment Statement,\u0027 \u0027Tax Reference,\u0027 and \u0027SA 100\u0027 further confirm this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.1392857142857143,
"features": {
"bank_statement": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.0,
"total_score": 0.10588235294117647
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.14285714285714285,
"total_score": 0.1392857142857143
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.030303030303030304,
"total_score": 0.05181818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 8,
"text_length": 2555
},
{
"confidence_score": 0.9,
"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, a tax adviser\u0027s name and address, and a client reference. Additionally, the mention of \u0027SA 100\u0027 suggests it is related to self-assessment tax returns, which aligns with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoh1yNq8ZWNJzXLHJrDI12WCz7F\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, a tax adviser\\\u0027s name and address, and a client reference. Additionally, the mention of \\\u0027SA 100\\\u0027 suggests it is related to self-assessment tax returns, which aligns with corporation tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294651, 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=1179, total_tokens=1399, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_toke"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information, including a tax reference number, a tax adviser\u0027s name and address, and a client reference. Additionally, the mention of \u0027SA 100\u0027 suggests it is related to self-assessment tax returns, which aligns with corporation tax documents.",
"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, a tax adviser\u0027s name and address, and a client reference. Additionally, the mention of \u0027SA 100\u0027 suggests it is related to self-assessment tax returns, which aligns with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.065,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "nominal_ledger"
}
},
"page_number": 9,
"text_length": 577
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. These are 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-DZFoiwqBmYvZn96ROjebb1ZLM7b8p\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to a tax return (SA 100), tax reference numbers, and declarations related to tax filings. These are 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=1777294652, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=198, prompt_tokens=1264, total_tokens=1462, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(audio_tokens=0, cached_tokens=1152)), prompt_filter_results=[{\u0027prompt_index\u0027: 0, \u0027co"
},
"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\": 0.9,\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": 1071
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and tax reporting, aligning with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFokQqWrd0VSERMceXECSxnVmCIX\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and tax reporting, aligning with corporation tax documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294654, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1566, total_tokens=1780, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(au"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and tax reporting, aligning with corporation tax documents.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax year, self-employment, and business income, which are key indicators of a tax-related document. Specifically, it appears to be related to self-employment and tax reporting, aligning with corporation tax documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.21741176470588236,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.23529411764705882,
"pattern_score": 0.1,
"total_score": 0.21741176470588236
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.15789473684210525,
"pattern_score": 0.0,
"total_score": 0.11368421052631578
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.0,
"total_score": 0.13
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 11,
"text_length": 1833
},
{
"confidence_score": 0.9,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data related to business expenses, including categories such as wages, rent, advertising, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFol1oHLJahwTT0EirBC4hcejwcf\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains detailed financial data related to business expenses, including categories such as wages, rent, advertising, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294655, 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=1586, total_tokens=1812, completion_tokens_details=CompletionTokensDetails(accepted_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 contains detailed financial data related to business expenses, including categories such as wages, rent, advertising, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains detailed financial data related to business expenses, including categories such as wages, rent, advertising, and professional fees. These are typical components of a financial statement, specifically an income statement or expense report. There are no indicators of transactions, loans, mortgages, or other document types listed in the classification categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.03529411764705882
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.06666666666666667,
"pattern_score": 0.0,
"total_score": 0.04
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 12,
"text_length": 1495
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to tax-related calculations, such as net profit or loss, capital allowances, taxable profit or loss, and specific tax-related boxes (e.g., box 47, box 48, box 63). These are indicative of a corporation tax document, likely part of 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-DZFomxN3DeDgY6WRPH09l4VfcmUTA\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 63). These are indicative of a corporation tax document, likely part of 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=1777294656, 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=1753, total_tokens=1980, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_toke"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to 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 63). These are indicative of a corporation tax document, likely part of a tax return or computation.",
"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 63). These are indicative of a corporation tax document, likely part of 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": 0.0757894736842105,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.10526315789473684,
"pattern_score": 0.0,
"total_score": 0.0757894736842105
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 13,
"text_length": 2581
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 68, box 73.3). It also mentions \u0027Self-employment (full) notes\u0027 and \u0027SA103F,\u0027 which are associated with UK self-assessment tax forms. 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-DZFooVGqv45XzO1gTZyleWukS31vY\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 68, box 73.3). It also mentions \\\u0027Self-employment (full) notes\\\u0027 and \\\u0027SA103F,\\\u0027 which are associated with UK self-assessment tax forms. 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=1777294658, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=241, prompt_tokens=1735, total_tokens=1976, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, re"
},
"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 68, box 73.3). It also mentions \u0027Self-employment (full) notes\u0027 and \u0027SA103F,\u0027 which are associated with UK self-assessment tax forms. 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 taxable profit or loss, tax years, adjustments for accounting periods, and specific tax-related boxes (e.g., box 68, box 73.3). It also mentions \u0027Self-employment (full) notes\u0027 and \u0027SA103F,\u0027 which are associated with UK self-assessment tax forms. 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": 0.08470588235294117,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.08470588235294117
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.05263157894736842,
"pattern_score": 0.0,
"total_score": 0.03789473684210525
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "corporation_tax"
}
},
"page_number": 14,
"text_length": 2614
},
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial statements.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFopKOYa3KkHa0118jh5UpLEyZxH\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital accounts. Key indicators include terms like \\\u0027Balance sheet,\\\u0027 \\\u0027Assets,\\\u0027 \\\u0027Liabilities,\\\u0027 \\\u0027Net business assets,\\\u0027 and \\\u0027Capital account,\\\u0027 which are characteristic of financial statements.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294659, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1545, total_tokens=1759, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), promp"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a balance sheet with detailed information about assets, liabilities, and capital accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial statements.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet with detailed information about assets, liabilities, and capital accounts. Key indicators include terms like \u0027Balance sheet,\u0027 \u0027Assets,\u0027 \u0027Liabilities,\u0027 \u0027Net business assets,\u0027 and \u0027Capital account,\u0027 which are characteristic of financial statements.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.2374736842105263,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.2631578947368421,
"pattern_score": 0.1,
"total_score": 0.2374736842105263
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "financial_statement"
}
},
"page_number": 15,
"text_length": 1436
},
{
"confidence_score": 0.9,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax-related information, such as \u0027Class 2 and Class 4 National Insurance contributions,\u0027 \u0027profits from self-employment,\u0027 and \u0027Tax Reference.\u0027 These are strong indicators of a tax-related document, specifically related to self-assessment or corporation tax. The presence of terms like \u0027SA103F\u0027 and \u0027Tax Reference\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZForwn5v0PsaAgTLIx74GeQl5cKR\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to tax-related information, such as \\\u0027Class 2 and Class 4 National Insurance contributions,\\\u0027 \\\u0027profits from self-employment,\\\u0027 and \\\u0027Tax Reference.\\\u0027 These are strong indicators of a tax-related document, specifically related to self-assessment or corporation tax. The presence of terms like \\\u0027SA103F\\\u0027 and \\\u0027Tax Reference\\\u0027 further supports this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294661, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=238, prompt_tokens=1211, total_tokens=1449, completion_tokens_details=CompletionTokensDetails(accept"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax-related information, such as \u0027Class 2 and Class 4 National Insurance contributions,\u0027 \u0027profits from self-employment,\u0027 and \u0027Tax Reference.\u0027 These are strong indicators of a tax-related document, specifically related to self-assessment or corporation tax. The presence of terms like \u0027SA103F\u0027 and \u0027Tax Reference\u0027 further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to tax-related information, such as \u0027Class 2 and Class 4 National Insurance contributions,\u0027 \u0027profits from self-employment,\u0027 and \u0027Tax Reference.\u0027 These are strong indicators of a tax-related document, specifically related to self-assessment or corporation tax. The presence of terms like \u0027SA103F\u0027 and \u0027Tax Reference\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "invoice"
}
},
"page_number": 16,
"text_length": 766
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains references to HMRC, tax calculations, Self Assessment, and specific tax-related terms such as \u0027Tax calculation summary,\u0027 \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 \u0027Student Loan repayment,\u0027 and \u0027Postgraduate Loan repayment.\u0027 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-DZFostYDv5Frk1QGVqKKd9cNeKabb\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 calculations, Self Assessment, and specific tax-related terms such as \\\u0027Tax calculation summary,\\\u0027 \\\u0027Class 2 NICs,\\\u0027 \\\u0027Class 4 NICs,\\\u0027 \\\u0027Student Loan repayment,\\\u0027 and \\\u0027Postgraduate Loan repayment.\\\u0027 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=1777294662, 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=1682, total_tokens=1908, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tok"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to HMRC, tax calculations, Self Assessment, and specific tax-related terms such as \u0027Tax calculation summary,\u0027 \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 \u0027Student Loan repayment,\u0027 and \u0027Postgraduate Loan repayment.\u0027 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 calculations, Self Assessment, and specific tax-related terms such as \u0027Tax calculation summary,\u0027 \u0027Class 2 NICs,\u0027 \u0027Class 4 NICs,\u0027 \u0027Student Loan repayment,\u0027 and \u0027Postgraduate Loan repayment.\u0027 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": 2275
},
{
"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 \u0027Blind person\u0027s surplus allowance,\u0027 \u0027married couple\u0027s surplus allowance,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document or related tax filing.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFouazEtfT1MoeVHTfXTU3dn1xOU\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 \\\u0027Blind person\\\u0027s surplus allowance,\\\u0027 \\\u0027married couple\\\u0027s surplus allowance,\\\u0027 and \\\u0027self-assessment statement of account.\\\u0027 These are strong indicators of a corporation tax document or related tax filing.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294664, 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=1329, total_tokens=1563, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to tax adjustments, self-assessment, tax years (2024-25, 2025-26), and specific tax-related terms such as \u0027Blind person\u0027s surplus allowance,\u0027 \u0027married couple\u0027s surplus allowance,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document or related tax filing.",
"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 \u0027Blind person\u0027s surplus allowance,\u0027 \u0027married couple\u0027s surplus allowance,\u0027 and \u0027self-assessment statement of account.\u0027 These are strong indicators of a corporation tax document or related tax filing.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.07058823529411765,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.02564102564102564,
"total_score": 0.02564102564102564
}
},
"type": "bank_statement"
}
},
"page_number": 18,
"text_length": 1362
},
{
"confidence_score": 0.95,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains tax-related information, including a tax reference number, National Insurance (NI) number, income and outgoings, tax rates, and a computation summary for the year of assessment. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Income Tax,\u0027 \u0027NIC,\u0027 and \u0027Taxable Income\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFovCWVsU05E1mLlcTWOnSnTFxI7\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 reference number, National Insurance (NI) number, income and outgoings, tax rates, and a computation summary for the year of assessment. These are key indicators of a tax return or corporation tax document. The presence of terms like \\\u0027Tax Return Schedules,\\\u0027 \\\u0027Income Tax,\\\u0027 \\\u0027NIC,\\\u0027 and \\\u0027Taxable Income\\\u0027 further supports this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294665, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=244, prompt_tokens=1280, total_tokens=1524, completion_tokens_details=CompletionTokensD"
},
"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 reference number, National Insurance (NI) number, income and outgoings, tax rates, and a computation summary for the year of assessment. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Income Tax,\u0027 \u0027NIC,\u0027 and \u0027Taxable Income\u0027 further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains tax-related information, including a tax reference number, National Insurance (NI) number, income and outgoings, tax rates, and a computation summary for the year of assessment. These are key indicators of a tax return or corporation tax document. The presence of terms like \u0027Tax Return Schedules,\u0027 \u0027Income Tax,\u0027 \u0027NIC,\u0027 and \u0027Taxable Income\u0027 further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.11666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.030303030303030304,
"total_score": 0.08181818181818182
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 19,
"text_length": 878
},
{
"confidence_score": 0.85,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. It does not include transaction details, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoxXvV8sDn8x4imJKRPiBtZlBHh\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. It does not include transaction details, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294667, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=214, prompt_tokens=1961, total_tokens=2175, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prom"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. It does not include transaction details, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains detailed financial data, including income, tax bands, and chargeable amounts, which are typical components of a financial statement. It does not include transaction details, account numbers, or specific identifiers related to invoices, loans, mortgages, or other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.06666666666666667,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.0,
"total_score": 0.06666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 20,
"text_length": 1708
},
{
"confidence_score": 0.85,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, investors\u0027 relief gains, and other financial metrics typically found in financial statements. There are no indicators of transactions, invoices, payroll, or property-related details, which rules out other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFoyFAQHCCV8NqyBOflJMxaBlQQD\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, investors\\\u0027 relief gains, and other financial metrics typically found in financial statements. There are no indicators of transactions, invoices, payroll, or property-related details, which rules out other categories.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294668, 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=1117, total_tokens=1334, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_predic"
},
"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, investors\u0027 relief gains, and other financial metrics typically found in financial statements. There are no indicators of transactions, invoices, payroll, or property-related details, which rules out other categories.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document contains references to capital gains tax (CGT), business asset disposal relief gains, investors\u0027 relief gains, and other financial metrics typically found in financial statements. There are no indicators of transactions, invoices, payroll, or property-related details, which rules out other categories.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 21,
"text_length": 402
},
{
"confidence_score": 0.9,
"document_type": "p32_payroll",
"features": {
"llm_based": {
"confidence": 0.9,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure and terminology align with tax obligations and payment schedules typically found in P32 payroll documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFp0XI51dk6JUYZFTfITLdsZlLdP\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"p32_payroll\",\\n \"confidence\": 0.9,\\n \"reasoning\": \"The document contains references to \\\u0027Client Payment Advice,\\\u0027 \\\u0027Payments Required,\\\u0027 \\\u0027Payment Dates,\\\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure and terminology align with tax obligations and payment schedules typically found in P32 payroll documents.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294670, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=221, prompt_tokens=1137, total_tokens=1358, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_predic"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure and terminology align with tax obligations and payment schedules typically found in P32 payroll documents.",
"raw_response": "```json\n{\n \"document_type\": \"p32_payroll\",\n \"confidence\": 0.9,\n \"reasoning\": \"The document contains references to \u0027Client Payment Advice,\u0027 \u0027Payments Required,\u0027 \u0027Payment Dates,\u0027 and specific payment amounts and dates, which are indicative of HMRC P32 Employer Payment Records. The structure and terminology align with tax obligations and payment schedules typically found in P32 payroll documents.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "p32_payroll"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 22,
"text_length": 400
},
{
"confidence_score": 0.85,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account.\u0027 These are indicative of a tax-related document, specifically related to individual or corporate tax obligations. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting, and the format resembles a tax computation or summary.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DZFp1zxm4O8zRRayzGTKtzkh42kT7\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"corporation_tax\",\\n \"confidence\": 0.85,\\n \"reasoning\": \"The document references tax-related terms such as \\\u0027Tax Due After Notional Taxes Paid,\\\u0027 \\\u0027Class 4 NIC Due,\\\u0027 and \\\u0027Payments on Account.\\\u0027 These are indicative of a tax-related document, specifically related to individual or corporate tax obligations. The mention of \\\u0027Year ended 5 April 2025\\\u0027 aligns with tax year reporting, and the format resembles a tax computation or summary.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1777294671, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=240, prompt_tokens=1111, total_tokens=1351, completion_tokens_details=CompletionTokensDetails(accepted_prediction_"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account.\u0027 These are indicative of a tax-related document, specifically related to individual or corporate tax obligations. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting, and the format resembles a tax computation or summary.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document references tax-related terms such as \u0027Tax Due After Notional Taxes Paid,\u0027 \u0027Class 4 NIC Due,\u0027 and \u0027Payments on Account.\u0027 These are indicative of a tax-related document, specifically related to individual or corporate tax obligations. The mention of \u0027Year ended 5 April 2025\u0027 aligns with tax year reporting, and the format resembles a tax computation or summary.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 23,
"text_length": 374
},
{
"confidence_score": 0.85,
"document_type": "corporation_tax",
"features": {
"llm_based": {
"confidence": 0.85,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\n \"reasoning\": \"The document 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 \u0027Total Class 4 NIC charge\u0027 suggests it is related to tax obligations, specifically for self-employed individuals. While it does not explicitly mention \u0027CT600\u0027 or \u0027corporation tax return\u0027, the content aligns most closely with the \u0027corporation_tax\u0027 category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFp372UnBUyjNe78yMxTh8UrS4Rh\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 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 \\\u0027Total Class 4 NIC charge\\\u0027 suggests it is related to tax obligations, specifically for self-employed individuals. While it does not explicitly mention \\\u0027CT600\\\u0027 or \\\u0027corporation tax return\\\u0027, the content aligns most closely with the \\\u0027corporation_tax\\\u0027 category.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294673, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=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 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 \u0027Total Class 4 NIC charge\u0027 suggests it is related to tax obligations, specifically for self-employed individuals. While it does not explicitly mention \u0027CT600\u0027 or \u0027corporation tax return\u0027, the content aligns most closely with the \u0027corporation_tax\u0027 category.",
"raw_response": "```json\n{\n \"document_type\": \"corporation_tax\",\n \"confidence\": 0.85,\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 \u0027Total Class 4 NIC charge\u0027 suggests it is related to tax obligations, specifically for self-employed individuals. While it does not explicitly mention \u0027CT600\u0027 or \u0027corporation tax return\u0027, the content aligns most closely with the \u0027corporation_tax\u0027 category.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "corporation_tax"
},
"method_used": "llm_preferred",
"pattern_based": {
"confidence": 0.03333333333333333,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.05555555555555555,
"pattern_score": 0.0,
"total_score": 0.03333333333333333
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.027777777777777776,
"pattern_score": 0.0,
"total_score": 0.03
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "invoice"
}
},
"page_number": 24,
"text_length": 359
},
{
"confidence_score": 0.7,
"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 tax-related document. However, it lacks specific indicators for other categories like invoices, payroll, or loan statements. The absence of transaction details, account numbers, or property-related information further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```",
"response_repr": "ChatCompletion(id=\u0027chatcmpl-DZFp5ixGyBAhrw9NRNVTZGx9Fi5tY\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 tax-related document. However, it lacks specific indicators for other categories like invoices, payroll, or loan statements. The absence of transaction details, account numbers, or property-related information further supports this classification.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null,\\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_period\": null,\\n \"mortgage_account_number\": null,\\n \"property_address\": null,\\n \"remaining_term_months\": null,\\n \"monthly_payment_amount\": null\\n }\\n}\\n```\u0027, refusal=None, role=\u0027assistant\u0027, annotations=[], audio=None, function_call=None, tool_calls=None), content_filter_results={\u0027hate\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027protected_material_code\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027protected_material_text\u0027: {\u0027detected\u0027: False, \u0027filtered\u0027: False}, \u0027self_harm\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027sexual\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}, \u0027violence\u0027: {\u0027filtered\u0027: False, \u0027severity\u0027: \u0027safe\u0027}})], created=1777294675, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=229, prompt_tokens=1057, total_tokens=1286, completion_tokens_details=CompletionTokensDetails(accepted_prediction_token"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document mentions \u0027Year ended 5 April 2025\u0027 and \u0027Personal Allowances,\u0027 which are indicative of a financial summary or tax-related document. However, it lacks specific indicators for other categories like invoices, payroll, or loan statements. The absence of transaction details, account numbers, or property-related information further supports this classification.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.7,\n \"reasoning\": \"The document mentions \u0027Year ended 5 April 2025\u0027 and \u0027Personal Allowances,\u0027 which are indicative of a financial summary or tax-related document. However, it lacks specific indicators for other categories like invoices, payroll, or loan statements. The absence of transaction details, account numbers, or property-related information further supports this classification.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null,\n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "uncertain",
"pattern_based": {
"confidence": 0.0,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "bank_statement"
}
},
"page_number": 25,
"text_length": 184
},
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and detailed financial data such as sales income, expenses, net profit, and capital allowances. These are key indicators of a financial statement, specifically an income statement for a sole proprietorship.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_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-DZFp66w5fnDURwY1vpNT4snll7Grz\u0027, choices=[Choice(finish_reason=\u0027stop\u0027, index=0, logprobs=None, message=ChatCompletionMessage(content=\u0027```json\\n{\\n \"document_type\": \"financial_statement\",\\n \"confidence\": 0.95,\\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and detailed financial data such as sales income, expenses, net profit, and capital allowances. These are key indicators of a financial statement, specifically an income statement for a sole proprietorship.\",\\n \"document_identifiers\": {\\n \"invoice_number\": null,\\n \"seller_name\": null, \\n \"invoice_date\": null,\\n \"total_amount\": null,\\n \"agreement_date\": null,\\n \"amount_financed\": null,\\n \"monthly_payment\": null,\\n \"loan_tenure\": null,\\n \"loan_reference\": null,\\n \"annual_interest_rate\": null,\\n \"outstanding_balance\": null,\\n \"statement_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=1777294676, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=212, prompt_tokens=1410, total_tokens=1622, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=Promp"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains a profit and loss account, tax adjustments, and detailed financial data such as sales income, expenses, net profit, and capital allowances. These are key indicators of a financial statement, specifically an income statement for a sole proprietorship.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a profit and loss account, tax adjustments, and detailed financial data such as sales income, expenses, net profit, and capital allowances. These are key indicators of a financial statement, specifically an income statement for a sole proprietorship.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.1616842105263158,
"features": {
"bank_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"corporation_tax": {
"keyword_score": 0.17647058823529413,
"pattern_score": 0.0,
"total_score": 0.12705882352941175
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.15789473684210525,
"pattern_score": 0.1,
"total_score": 0.1616842105263158
},
"hp_agreement": {
"keyword_score": 0.038461538461538464,
"pattern_score": 0.0,
"total_score": 0.023076923076923078
},
"invoice": {
"keyword_score": 0.16666666666666666,
"pattern_score": 0.0,
"total_score": 0.09999999999999999
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"nominal_ledger": {
"keyword_score": 0.08333333333333333,
"pattern_score": 0.0,
"total_score": 0.065
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "financial_statement"
}
},
"page_number": 26,
"text_length": 1304
},
{
"confidence_score": 0.95,
"document_type": "financial_statement",
"features": {
"llm_based": {
"confidence": 0.95,
"features": {
"api_response_debug": {
"content_preview": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.95,\n \"reasoning\": \"The document contains a balance sheet summary with assets, liabilities, and net business assets, as well as a capital account and profit/loss adjustments. These are key indicators of a financial statement, specifically a balance sheet and income 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-DZFp8YcyvLGfrxPdRegBNREjsCqkS\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 capital account and profit/loss adjustments. These are key indicators of a financial statement, specifically a balance sheet and income 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=1777294678, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=209, prompt_tokens=1378, total_tokens=1587, completion_tokens_details=CompletionTokensDetails(accepted_prediction_tokens=0, audio_tokens=0, reasoning_tokens=0, rejected_prediction_tokens=0), prompt_tokens_details=PromptTokensDetails(a"
},
"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 capital account and profit/loss adjustments. These are key indicators of a financial statement, specifically a balance sheet and income 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 capital account and profit/loss adjustments. These are key indicators of a financial statement, specifically a balance sheet and income statement.\",\n \"document_identifiers\": {\n \"invoice_number\": null,\n \"seller_name\": null, \n \"invoice_date\": null,\n \"total_amount\": null,\n \"agreement_date\": null,\n \"amount_financed\": null,\n \"monthly_payment\": null,\n \"loan_tenure\": null,\n \"loan_reference\": null,\n \"annual_interest_rate\": null,\n \"outstanding_balance\": null,\n \"statement_period\": null,\n \"mortgage_account_number\": null,\n \"property_address\": null,\n \"remaining_term_months\": null,\n \"monthly_payment_amount\": null\n }\n}\n```"
},
"type": "financial_statement"
},
"method_used": "both_agree",
"pattern_based": {
"confidence": 0.25768421052631574,
"features": {
"bank_statement": {
"keyword_score": 0.11764705882352941,
"pattern_score": 0.0,
"total_score": 0.07058823529411765
},
"corporation_tax": {
"keyword_score": 0.058823529411764705,
"pattern_score": 0.0,
"total_score": 0.042352941176470586
},
"credit_card": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"financial_statement": {
"keyword_score": 0.15789473684210525,
"pattern_score": 0.3,
"total_score": 0.25768421052631574
},
"hp_agreement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"invoice": {
"keyword_score": 0.1111111111111111,
"pattern_score": 0.125,
"total_score": 0.11666666666666667
},
"loan_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"mortgage_statement": {
"keyword_score": 0.02631578947368421,
"pattern_score": 0.0,
"total_score": 0.025263157894736838
},
"nominal_ledger": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"p32_payroll": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
},
"payroll_statement": {
"keyword_score": 0.013888888888888888,
"pattern_score": 0.0,
"total_score": 0.015
},
"property_statement": {
"keyword_score": 0.0,
"pattern_score": 0.0,
"total_score": 0.0
}
},
"type": "financial_statement"
}
},
"page_number": 27,
"text_length": 1111
},
{
"confidence_score": 0.7,
"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 contains information about residence status, domicile, and dual residence status for a specific year, which aligns with personal or financial reporting. However, it lacks specific indicators for other categories such as transactions, invoices, payroll, or loan details. The mention of \u0027Year ended 5 April 2025\u0027 suggests it could be part of a financial or tax-related document, but the absence of explicit tax forms or financial statement elements 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-DZFp9g3jPiUjLYxdQyUVqSneALGoJ\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 contains information about residence status, domicile, and dual residence status for a specific year, which aligns with personal or financial reporting. However, it lacks specific indicators for other categories such as transactions, invoices, payroll, or loan details. The mention of \\\u0027Year ended 5 April 2025\\\u0027 suggests it could be part of a financial or tax-related document, but the absence of explicit tax forms or financial statement elements 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=1777294679, model=\u0027gpt-4o-2024-11-20\u0027, object=\u0027chat.completion\u0027, service_tier=\u0027default\u0027, system_fingerprint=\u0027fp_af7f7349a4\u0027, usage=CompletionUsage(completion_tokens=25"
},
"document_identifiers": {
"agreement_date": null,
"amount_financed": null,
"annual_interest_rate": null,
"invoice_date": null,
"invoice_number": null,
"loan_reference": null,
"loan_tenure": null,
"monthly_payment": null,
"monthly_payment_amount": null,
"mortgage_account_number": null,
"outstanding_balance": null,
"property_address": null,
"remaining_term_months": null,
"seller_name": null,
"statement_period": null,
"total_amount": null
},
"llm_reasoning": "The document contains information about residence status, domicile, and dual residence status for a specific year, which aligns with personal or financial reporting. However, it lacks specific indicators for other categories such as transactions, invoices, payroll, or loan details. The mention of \u0027Year ended 5 April 2025\u0027 suggests it could be part of a financial or tax-related document, but the absence of explicit tax forms or financial statement elements reduces confidence slightly.",
"raw_response": "```json\n{\n \"document_type\": \"financial_statement\",\n \"confidence\": 0.7,\n \"reasoning\": \"The document contains information about residence status, domicile, and dual residence status for a specific year, which aligns with personal or financial reporting. However, it lacks specific indicators for other categories such as transactions, invoices, payroll, or loan details. The mention of \u0027Year ended 5 April 2025\u0027 suggests it could be part of a financial or tax-related document, but the absence of explicit tax forms or financial statement elements 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": 28,
"text_length": 327
}
],
"processing_timestamp": "2026-04-27T12:58:01.302123"
}