This whitepaper offers a comparison of Paradatec’s AI-based full-page textual analysis methodology with the approaches by other offerings in the mortgage processing automation market.


The objective of this paper is to highlight the differences in methodology between the document recognition technology offered by Paradatec and that offered by other vendors. There is no intention to identify specific technologies by name. We hope that the comparison will allow the reader to make more informed decisions about the environments that may be more suitable for each methodology.

High-Level Approach

The most fundamental difference in approach between Paradatec and the majority of other “advanced” document recognition technologies is that Paradatec treats variable layout documents as unstructured documents whereas most other prominent solutions treat them more as semi-structured documents. To illustrate the difference in methodology let us consider a customer wishing to process pay stub documents. A Paradatec implementation would typically be deployed with one set of completely generic rules designed to encompass all variations of the “Pay Stub” document type from any company. The software is capable of being configured to perform conditional processing for specific exceptions to our generic rules (per-layout exceptions) but it is not typically necessary to do this.


You can download the full whitepaper here: