News story

Hypothesis generation and visualisation from big data

Case study from the University of Sheffield who pitched their ideas to industry and investors at the CDE Marketplace on 6 September 2016.

HiDE

The volume of electronic data now available presents a problem for anyone trying to track an area of interest.

The University of Sheffield has been funded by the Centre for Defence Enterprise (CDE) to see how literature-based discovery (LBD), which is a text-mining technology, could automatically analyse large volumes of information and identify connections that can only be made by combining information from multiple sources.

It has developed HiDE (Hidden Discovery Explorer), and LBD system which makes use of complex mathematical techniques to allow more information to be analysed than possible previous approaches. HiDE’s output is displayed visually and can be viewed in a web browser.

HiDE was originally developed as part of a project funded by the Engineering and Physical Sciences Research Council (EPSRC) in which large repositories of life-science publications (containing more than 26 millions documents) were processed and the output analysed by doctors to identify information that was relevant to neuroscience and oncology.

The aim of this CDE project was to apply HiDE to scenarios that are relevant to defence and security. The project tested HiDE on a range of relevant document collections including collections of articles on defence related topics (weapons, terrorism, defence and military personnel) and a collection of documents about military hardware.

The University has developed a range of strategies to conduct intelligent text analysis, application of inference rules to filter out hidden knowledge and identifying unhelpful ‘linking terms’. It plans to develop a version of HiDE which allows users to choose individual filtering strategies, thereby providing a personalised view of the information generated.

The University’s Dr Mark Stevenson says:

The funding we received from CDE allowed us to develop our approach beyond its original application and demonstrate that it could be applied to documents on any topic.

View the pitch presentation slides.

About CDE

CDE funds novel, high-risk, high-potential-benefit research. We work with the broadest possible range of science and technology providers, including academia and small companies, to develop cost-effective capabilities for UK armed forces and national security.

CDE is part of Dstl.

Centre for Defence Enterprise

Building R103
Fermi Avenue
Harwell Oxford
Oxfordshire
OX11 0QX

Email cde@dstl.gov.uk

Telephone +44 (0)30 67704236

Alternative number +44 (0)30 67704237

Please email for the quickest response.

Published 15 August 2016