TNA upgrades archive searches

The tool will use text mining and semantic annotation which models the relationships between words and concepts, TNA told GC News. It will link, or infer linkages between, similar concepts and provide for more intuitive searches of archived UK government websites. These currently require the person carrying out the search to know which department was responsible for particular functions. "If you want to know the health implications of swine flu you would go to NHSDirect or the Department of Health. But if you wanted to understand its economic impact you would need to go to treasury.gov.uk or Business Link," said a TNA spokesperson. Furthermore, in a web archive, Google type searching is not as effective as it is in a live web context. It is also limited as it does not take into account changes over time. "To address these problems, The National Archives, in partnership with a consortium of suppliers, is researching and analysing innovative new technology that will help link data over time," said the spokesperson. "This approach aims to free up more information in this web archive resource to those who do not necessarily have an understanding of the formal structure of government." The tool will be developed by a consortium of System Simulation, the GATE team at University of Sheffield and Bulgaria based Ontotext, which was awarded the contract for the work on 22 June 2010.