Case Study: a regulatory technology provider achieves 80%+ document classification automation with Aluma

Mortgage and Loans Intelligent Data Extraction and Automation

Large Technology Provider, a regulatory technology provider serving the mortgage industry, was processing more than 120 million documents a year and needed a better way to handle over 600 highly variable mortgage document types. They turned to Aluma and its intelligent data extraction and automation platform to improve document classification and data extraction at growing scale.

Aluma applied a combination of machine learning, NLP, and rules-based techniques to train and test the model on thousands of mortgage documents in just minutes, with no manual tuning. The result was over 80% automation for document classification at 99.5%+ accuracy and extraction of more than 88% of required data elements, helping reduce costs, improve data quality, and support processing of over one billion pages per year in 2022.