Content Analytics Toolset
In partnership with Aston University, we are the lead in a breakthrough R&D project entitled 'CAT' for Content Analytics Toolset.This involves semantic analysis of record and document content to a degree not currently available.
The CAT project seeks to analyse the contents (unstructured data) of large corpuses of documents and documentary records to generate semantically relevant content topics and themes within context. And it combines expertise in machine learning, statistical modelling & text mining, probabilistic mathematics & natural language processing, data exploration, cognitive design, document management & paper digitisation, and new approaches to very advanced OCR with programmatic metadata capture & extraction.
This project was generously grant funded by Innovate UK (prev. Technology Strategy Board - www.innovateuk.gov.uk). The official title of the project was 'Big Data: Digitisation, Semantic Analysis, Topic Modelling, Visualisation and Exploration'. And the current output is a first stage Content Analytics Toolset (CAT/1).
All CAT/1 project goals were successfully achieved as we developed a new algorithm for handwritten text detection, an integrated framework for topic extraction from digitised (paper) patient records, and a topic visualisation system for navigation and retrieval of semantically relevant documents.
CAT/1 is now in product development particularly regarding data redaction of sensitive documents (Central Government & Parliamentary Institutions) and metadata extraction (Health & Social Care Records). Pilot feedback is very positive and has opened larger opportunities that require more extensive & expansive R&D building on from CAT/1 which, not unsurprisingly, is named CAT/2.
CAT/2 will accelerate us to a new state-of-the-art where we unlock the semantic meaning lost within unstructured data wherever it exists in any Line of Business data system via an entirely new approach to Text Analysis that is not currently available anywhere in the world.