The campaign to rid the mortgage industry of as many cumbersome, expensive, and error-prone manual processes as possible continues as AI Foundry unveils the latest enhancement to its Cognitive Business Automation Platform. The company announced this week that its new mortgage document model will add key functionality to its automation platform, leveraging machine learning, machine vision, and AI to provide better automated data classification and data extraction.
“The model enables any lender to upload its loan application material and in return receive fully indexed and extracted data within seconds,” Head of Solution Development for AI Foundry Peter Piela explained. Piela added that the technology has a 95% accuracy rate, which he said was comparable to manual accuracy rates, and is superior to legacy text classification methods. Leveraging both cognitive and deep neural techniques, the solution was trained on more than 100,000 mortgage documents, 300 document types, and 2,000 data extractions.
“The impact of using our document model means significant time savings for the lender and the replacement of expensive manual processes with far more efficient automated ones,” Piela said.
And because the mortgage document model is part of AI Foundry’s Cognitive Business Automation Platform, users will be able to work with the existing model with its powerful rules engine, as well as broaden the functionality of the platform to manage specific mortgage workflow processes. In its statement, the company noted that the model is enhanced “continuously” with new variants that are easily deployed and equally accessible to customers thanks to the solution’s SaaS environment.
Headquartered in Wakefield, Massachusetts, AI Foundry demonstrated its Agile Mortgage solution at FinovateFall last year. In February, AI Foundry teamed up with fellow Finovate alum Ellie Mae to put AI to work accelerating the mortgage lending process. Steve Butler is founder and President of the company, which was founded in 2016.