Classifying and Extracting Mortgage Loan Data with Amazon Textract

Mortgage loan applications, at least in the United States, comprise around 500 or more pages of diverse documents. In order for applications to be reviewed, all these documents need to be classified, and the data on each form extracted. This isn't as easy as it might sound! Besides different data structures in each document, the same data element may have different names on different documents--for example, SSN, or Social Security Number, or Tax ID. These three all refer to the same data.