Taxonomic information retrieval (TAXIR) from the World Wide Web: Knowledge-based query and results refinement with user profiles and decision models

The purpose of this research is to establish a methodology for selecting the best solution from available candidates. Currently, the Internet and the World Wide Web provide a laboratory for this research. The Web also allows access to a great deal of information about a vast array of subjects. A user can begin a search for information by selecting a Web page and following the embedded links from page to page looking for clues to the desired information. An alternative method is to use one of the Web-based search engines to select the Web pages that refer to the general subject of the information desired. In either case, a vast amount of information is retrieved. The quantity can be overwhelming, and much of the information may be irrelevant to the user's needs.; We present a methodology for query construction and results analysis that provides the user with a ranking of choices based on the user's determination of importance. The query is initially designed by the user with assistance from the user's profile, a thesaurus, and previously constructed queries as a taxonomy of the information need. After the query has returned its results, decision analytic methods and information source reliability knowledge are used on the expanded taxonomy to rank the solution candidates.