Skip to main content

Data Extraction

Finovox can automatically extract structured data from documents. This includes information like names, dates, amounts, identification numbers, and much more depending on the document type.

Document Quality

The accuracy and completeness of data extraction depends on the quality of the document. Clear, high-resolution documents with readable text will yield better extraction results.

Available Fields​

Below is the complete list of fields that Finovox can extract from documents:

Loading extracted fields...

API Usage​

To request data extraction, add "data_extraction" to the analyse_type in your payload:

{
  "analyse_type": ["data_extraction"]
}

Extracted data is returned in the information field:

{
  "information": {
    "invoice_number": "123456",
    "issue_date": "2023-01-25",
    "total_amount": "957.93",
    "issuer_name": "Company Name",
    "recipient_name": "Client Name"
  }
}

Field Availability​

Not all fields are available for all document types. The fields extracted depend on:

  • The document category (tag)
  • The information present in the document