Extract is an Intelligent Reading tool that can read, understand and derive information from complex, unstructured data in any language or format. Using our proprietary Cleansing, Classification, Optical Character Recognition and Natural Language Processing capabilities, Omnitive Extract works well with both Hard-Copy (Scanned or Photo) and Soft-Copy documents. It processes heaps of document types to extract information that is most relevant to your business. In the past, we have process simple, Structured Documents (Ex. Passports), Semi-Structured Documents (Ex. Invoices), and Unstructured Documents (Ex. Annual Reports or Legal Documents) with >80% accuracy.
This solution helps businesses simplify their onboarding, identification and authentication needs as part of your end-to-end process. For example, such a model fits in perfectly with Account Opening at a Bank, Processing any type of Visa application, etc.
The system is trained to extract the data points from a Passport in English language, if the data is present on the document. TAIGER can make the same model available in Spanish or Russian language, please feel free to reach out to us.The following are the data points that the system is trained to extract:
Date of Birth
Should you encounter any empty fields or “NA”, the following are some of the possibilities:
Data point is not present on the document, masked/covered, noise (caused by handwritten text, stamps, etc).
The model is not trained based on your sample document
Note: Unlike most capturing software out there, we do not use templates. The system is trained based on variances - the more varying documents the system is trained on, the more accurately it can extract any varying types of the document.
Reach out to the TAIGER team if you encounter any issues with the extraction or if you wish to understand how our technology works.