The goal of this project is to ensure the smooth extraction of insurance-related entities from unstructured data. The unstructured data is to be provided as input which is obtained from any of the OCR engines by processing their documents in the image of PDF formats.
This AIFabric compatible ML model analyses incoming insurance documents and extracts entities i.e. policy number, policy tenure & client name and the extracted data can be utilized by the bot to perform data entry activities to their respective application.
The model has been trained on the client and carrier-specific insurance documents compatible with the Accord template.
1. Download the zip file from drive and upload the package to AIFabric
2. Input for the model is string obtained from OCR(Preferably tesseract) after passing the insurance document.
3. Output will be json object with keys "POLICY_NO","POLICY_PERIOD","INSURED_ORG".