Classify

The Classify feature of Ocrolus is designed to identify and categorize PDFs and images into specific form types, including tax forms, ID proofs, bank statements, and pay stubs. By grouping similar form types, it also performs document collation. With document classification, you can quickly get an overview of the contents of a loan file or a mixed document, enabling you to identify whether all the required forms are present. This feature helps you avoid the complexity of exception handling that often arises when uploading miscategorized documents to other endpoints. To see the complete list of supported form types, see forms directory.

The end-to-end flow of document classification using Ocrolus' classification feature involves several steps. You can classify the following types of documents using Classify:

The example below shows the JSON response of the classification summary returned after the document is classified.

{
  "status": 200,
  "message": "OK",
  "response": {
    "forms": [
      {
        "form_uuid": "71cb2e19-cf3b-4a1e-91c7-7b1703513491",
        "upload_details": {
          "mixed_doc_uuid": "b41ca80c-a4d1-4373-9329-90e7e5aba92d",
          "mixed_doc_page_indexes": [...]
        }
      },
      {
        "form_type": {...},
        "upload_origin": "MIXED"
      }
    ],
    "mixed_doc_uuid": "b41ca80c-a4d1-4373-9329-90e7e5aba92d"
  }
}