Development of an Intelligent Search Engine for Knowledge Resources

Traditional knowledge sources, such as thesauri and subject headings lists, are helpful for manual use, but the format of the encoded semantic information and the prevalent indexing and search tools are not well suited for Web retrieval applications. Classic examples of such knowledge sources are the ETDE/INIS Joint Thesaurus for energy information, the Library of Congress Subject Headings, and the National Library of Medicine¿s MeSH (Medical Subject Headings) Thesaurus. This project will develop an intelligent search engine prototype to find the best potentially-applicable subject headings for arbitrary multi-concept search queries. The new Concept Search pilot will be integrated into an advanced meta-search and discovery engine to support enhanced query refinement. The effectiveness of the new thesaurus indexing and retrieval algorithms will be tested and evaluated in the context of searching large energy-related scientific and technical databases. Commercial Applications and other Benefits as described by the awardee: The use of rich semantic information identified by the intelligent Concept Search Engine is expected to improve the quality of search results. In addition to energy-related applications, the Concept Search Engine should be applicable to other scientific and technical knowledge sources and retrieval applications.