An expert system for automatic query reformation

Susan Gauch, John B. Smith
1993 Journal of the American Society for Information Science  
Unfamiliarity with search tactics creates difficulties for many users of online retrieval systems. User observations indicate that even experienced searchers use vocabulary incorrectly and rarely reformulate their queries. To address these problems, an expert system for online search assistance was developed. This prototype automatically reformulates queries to improve the search results, and ranks the retrieved passages to speed the identification of relevant information. Users' search
more » ... nce using the expert system was compared with their search performance on their own, and their search performance using an online thesaurus. The following conclusions were reached: 1) The expert system significantly reduced the number of queries necessary to find relevant passages compared with the user searching alone or with the thesaurus. 2) The expert system produced marginally significant improvements in precision compared with the user searching on their own. There was no significant difference in the recall achieved by the three system configurations. 3) Overall, the expert system ranked relevant passages above irrelevant passages. INTRODUCTION Driving Problem Technology to produce, store, and distribute massive quantities of electronic information has matured. Textbases, online full-text databases, are being created in Gauch -2many fields. Personal workstations have become common. To make use of the information accessible from their desks, technology must be developed which allows end-users to search effectively. One user study found that whereas system mechanics are rarely a problem for any but very inexperienced and infrequent users, even experienced searchers have significant problems with search strategy and output performance (Borgman, 1986a). Another found experienced searchers lost sight of the search logic, missed obvious synonyms, and searched too simply (Fenichel, 1981) . In spite of low recall, half of the searchers never modified the original query in an attempt to improve their results. Studies of inexperienced searchers find even more problems with search strategy. In one study, a quarter of the subjects were unable to pass a benchmark test of minimum searching skill (Borgman, 1986b) . In an experiment contrasting the searching of novices versus experienced searchers, the novices found some relevant documents easily, but they failed to achieve high recall and were unable to reformulate queries well (Oldroyd, 1984) . The experienced searchers in this study were more persistent and willing to experiment than the novices. Blair and Maron (1985) paint an even bleaker picture for searching full-text databases. Legal assistants searching a legal database achieved only 20% recall, although they were attempting to do a high recall search. The factors, as identified by the authors, leading to this poor performance were poor searching technique (failure to use stemming and synonyms), stopping the query iteration too soon, and the inability to search on inter-document relationships. The authors argued that vocabulary problems make high recall impossible on full-text databases.
doi:10.1002/(sici)1097-4571(199304)44:3<124::aid-asi2>3.0.co;2-c fatcat:rutler5tubh6ddn44ax32hrvr4