Automatic Summarisation for Systematic Reviews using Text Mining
This project (ASSERT) centres on developing an exemplar summarisation service for the social science domain in the form of a case study of systematic reviews for the Evidence for Policy and Practice Information Centre (EPPI). The project also features a community call around the social science domains to develop further case studies showing the benefits of text mining. Through all activities the project focuses on providing for broader institutional involvement in text mining.
Aims and objectives
The overall aim of ASSERT is to encourage greater participation by the social science community in e-Research by developing text mining technology to facilitate the production of systematic reviews and to support a number of community projects related with text mining applications. Our overall objectives are to:
- liaise with domain experts who will use text mining as an integral part of their projects under the community call
- configure appropriate text mining workflows for the specific purposes of the supported projects
Following a requirements analysis process we will customise existing text mining tools for lexico-syntactic annotation, term discovery, named entity recognition, clustering and classification to meet the needs of the social science researchers. This will be integrated with a newly developed summarisation service to act as assistive technologies for carrying out systematic reviews. These will be evaluated with real challenges as part of the case studies before being released to the community.
Anticipated outputs and outcomes
We anticipate project outputs to include an exemplar service for summarisation along with appropriate documentation for users and additional training support material. As this service, and the additional call projects, build upon a number of core technologies we hope to deliver smaller services representing the components that make up this workflow. Finally we will organise a workshop bringing together key participants and stakeholders with future potential users to demonstrate the benefits of using text mining. We anticipate an improvement in awareness and enhanced take-up of text mining in the social sciences.
Technology / Standards used
The project draws upon core NaCTeM (National Centre for Text Mining) technology and integrates this with further functionality from external open source tools. Best practice will be maintained throughout the project with appropriate standards being adhered to allowing for maximum interoperability with other services and repositories.
- National Centre for Text Mining (NaCTeM), University of Manchester
- Mr. Brian Rea, University of Manchester, School of Computer Science, Tel: 0161 306 3094, Fax: 0161 306 5201 firstname.lastname@example.org