The four Systemic Banks needed an intelligent solution to achieve data extraction from complex, multi-page Legal Documents. The legal documents were created by thousands of different lawyers and were fully unstructured, using variations in phrasing and document layouts.
The back-end Robotic Process Automation (RPA) Data-Extraction technology that was implemented, used a variety of proximity algorithms to locate specific phrases and words before extracting the required metadata from the area of interest. The Project required the extraction of more than 270 field metadata from hundreds of thousands of legal documents.