JISC

Overview

Text mining, that interrogates and sorts thousands of electronic documents, has the potential to revolutionise the identification of research and collective expertise for policy-making and practice. This project will develop and test text mining to support efficient access to evidence through a web portal. Currently, access to evidence is either available through general web search engines or specialist bibliographic databases. The scope of bibliographic databases is often narrow, requiring many searches, whereas the scope of a search engine such as Google is very broad, resulting in huge numbers of irrelevant hits.

Aims and objectives

We aim to evaluate an innovative search engine – using text mining – for a portal of education evidence, relevant to education practitioners and policy-makers.

Project methodology

Requirements analysis We will work with the current UK Education Evidence Portal partners to identify current users of the portal such as teachers, information specialists and local government policy-makers. Their requirements of the portal will then be assessed using online questionnaires and focus groups. NaCTeM will then develop the text mining search engine based on these requirements.

Comparison We will conduct a comparison of three variations on how to search the UK Educational Evidence Portal following which the search engine will be refined.

User-testing This which will include one-to-one sessions with users which will be observed and followed up with interviews.

Anticipated outputs and outcomes

• A text mining search engine incorporated into the current search options of the UK Evidence for Education Portal
• The project will be written up as a case study in using text mining tools, and published in a peer reviewed journal and the project publicised via conferences and appropriate JISC events, and on EEP and consortium websites
• In line with current NaCTeM practice, the text mining tools themselves will be made available freely for others to use
• If successful, this project would be a high profile exemplar of the utility of text mining in the social sciences, with application beyond the single case described here

Lead institution

• Evidence for Policy and Practice Information and coordinating Centre (EPPI-Centre), Social Science Research Unit, Institute of Education, University of London 

Project partners

• The National Centre for Text Mining (NaCTeM)