Beyond Recognition offers terabyte text trial

Guided ‘visual classification’ methodology for document collection rationalization.

At the 2014 Pipeline Conference in Houston last month Beyond Recognition (BR) announced the offer of a free trial of its ‘visual classification’ approach to the management of oil and gas documents. Visual classification (VC) performs a graphical analysis of documents, grouping scanned documents with native-format files. Documents are managed using a normalized schema independent of their containers that allows for a single workflow spanning both paper and electronic documents.

VC starts out with a completely automated first pass classification that places files and scans into groups of visually-similar documents. Knowledge workers then perform checks on a few documents to evaluate document importance and to verify processing workflows. After a few weeks, BR claims that almost all incoming documents are placed in appropriate clusters.

A similar process is used to pull data elements from document clusters using information prepared by domain experts. Output formats are client-specific but typical deliverables are CSV files of normalized attributes, image-over-text PDFs and a single-instance of the original native file. The process results in a reduction of around 90% on a typical document collection. BR is offering a proof of concept trial on a terabye of corporate data. More from Beyond Recognition.

Click here to comment on this article

Click here to view this article in context on a desktop

© Oil IT Journal - all rights reserved.