Black Knight Continued Innovation with Document Extraction using AWS Machine Learning Capabilities

Black Knight, Inc. (NYSE:BKI) is transforming the mortgage industry and reimagining the home ownership journey for its customers by utilizing digital engagement, automation, and machine learning (ML). Black Knight Origination Technologies leverages AWS ML solutions to better standardize and validate bank statements, thereby increasing accuracy, reducing manual effort, and lowering costs.

This is an example of the constant effort that Black Knight is taking to provide innovative, high-performance software, data, and analytics for mortgage and home equity lending and servicing,

“Black Knight is focused on using the power of artificial intelligence (AI) to create innovative solutions that address the needs of the mortgage industry. Our AI product, called AIVA, drives efficiency across the Black Knight ecosystem by eliminating stare-and-compare work and by applying ML and advanced business rules to lower loan processing costs. The Black Knight ML team worked with the Amazon Machine Learning Solutions Lab to provide faster and more reliable processing behind Black Knight’s mortgage origination solutions, allowing our lender clients to be more competitive and efficient, providing operating leverage to further accelerate their growth.” Rich Gagliano, President, Origination Technologies.

Improving Accuracy and Reducing Cost in Document Extraction

Black Knight provides solutions to the mortgage industry that reduce manual processes, support compliance and quality, mitigate risk, and deliver significant cost savings to their clients, including many of the largest mortgage lenders in the United States. Due to the increasing complexity of regulatory requirements and changes in the mortgage loan lifecycle, sophisticated data and analytics supports each process. As the mortgage industry grows in complexity, Black Knight helps clients improve and scale their business processes with automation and an end-to-end solutions ecosystem that supports the entire real estate and mortgage lifecycle. Black Knight has developed integrated, AI/ML-enabled solutions and continuously strives for innovation to better serve clients.

Designing a Company-Driven Solution

Amazon Machine Learning Solutions Lab collaborated with Black Knight to augment Black Knight’s innovative AI/ML-powered solution to improve document extraction automation. They leveraged new features from Amazon Textract, and developed custom post-processing modules to improve the accuracy of data extraction from bank statements. Amazon Textract is an AI/ML service that extracts text, handwriting, and other forms of metadata from scanned documents, forms, and tables.

The new solution can identify transaction tables from differing bank statement layouts, and then convert them into a standard format for faster analysis and data parsing. Extra metadata in bank statements, such as the period start and end date, are extracted using domain-specific queries with the new Amazon Textract Queries function. The solution supports an automated checksum test to validate that the transaction table entries are correctly extracted, which then reports this accuracy number to downstream applications.

Continuing Innovation with AI/ML

Through this ML-powered solution for bank statement table extraction and validation, Black Knight expects a reduction in the time it takes to process bank statements, as well as a reduction in the overall number of bank statements passing through humans for processing. This process is repeatable for other documents and use cases within the mortgage domain.

The adoption of AWS Cloud technologies and capabilities has helped Black Knight address the growing complexity of the entire real estate-mortgage continuum. Using AWS, the company is continuing to advance applications for AIVA throughout the mortgage lifecycle, thereby helping lenders become more efficient by lowering loan-processing costs and providing serverless computing.