Issue No.04 - July/August (1999 vol.14)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/5254.784085
This article describes a system called T?traFusion that allows a kind of information seeking called information discovery to be supported. Information discovery involves harvesting a subset of resources from some larger collection, filtering this subset, and performing a data-mining operation on it. This procedure creates a reduced information structure that users can navigate to discover information about some domain. The system the authors have developed operates on the global World Wide Web and lets users specify a reduced domain in which to do information discovery via an initial request for information. In this way, the system can harvest a collection of Web pages that it then further mines and filters to turn this raw information into useful but unknown patterns automatically induced from an analysis of the pages' content. The results of the data-mining process are displayed graphically for the user to browse interactively.
Francis Crimmins, Alan F. Smeaton, Taoufiq Dkaki, Josiane Mothe, "TétraFusion: Information Discovery on the Internet", IEEE Intelligent Systems, vol.14, no. 4, pp. 55-62, July/August 1999, doi:10.1109/5254.784085