Parascript Brings New Features to FormXtra
Parascript, a provider of data extraction and interpretation powered by machine learning, has unveiled new capabilities in FormXtra 6.3.
The new release resolves the issues that customers face associated with multi-channel document capture without increasing complexity, said Greg Council, VP of marketing and product management at Parascript, noting that the focus is on meeting high-volume, complex document processing needs with flexibility for both processing rules and workflows regardless of the source.
According to the company, dealing with the large variance in image quality in documents resulting from different points of capture makes it difficult to achieve high-levels of automation. With FormXtra 6.3, the company says, Parascript enables companies to deal with any document, all within the same workflow with levels of performance that previously could only be observed with controlled high-quality scanners. Black and white forms such as claims can be converted to simulate drop-out ink forms; images are analyzed, and those that vary in size are reformatted to conform to expected layouts. Mobile images are transformed into high-quality scans—even without specialized mobile apps.
Parascript has also implemented Virtual Drop-out for Claims documents that results in improved classification and recognition solution for black-and-white claims, regardless of image quality or scale. The new claims module provides better classification and improves drop-out and black-and-white claims recognition.
FormXtra supports identification of electronic PDFs and can extract and parse the text leading to highly accurate results and greater throughput, and also enables documents to be organized and tested leveraging the text-based classifier, image classifier or a combination. While text is frequently used as the primary input in most capture systems, Parascript software can leverage both visual analysis of documents and text to accurately classify documents.
Using new deep-learning technology, invoice data location for header-footer data is now better than 80% with 95% accuracy at the field level, requiring no external validation data source and no manual validation necessary. Additionally, line item location is now better than 55% with 98% accuracy with no PO matching requirement.
In addition to out-of-the-box capabilities, FormXtra’s .NET API provides control over forms processing definitions, routines and output, giving customers a single solution that can process many types of business documents so that extracting key data is both straightforward and scalable. Dual support for both document-centric and field-centric data validation ensures that businesses can take advantage of sophisticated load-balancing, performance monitoring, automation and field-level security to support a wide variety of data capture needs.
Visit Parascript for more information.