An Algorithm to use Feedback on Viewed Documents to Improve Web Query: Enabling naive searchers to search the Web smartly

摘要

This paper presents an algorithm to improve a web search query based on the feedback on the vieweduddocuments. A user who is searching for information on the Web marks the retrieved (viewed) documents asudrelevant or irrelevant to further expose the information needs expressed in the original query. A new webudsearch query matching this improved understanding of the user's information needs is synthesized fromudthese text documents. The methodology provides a way for creating web search query that matches theuduser's information need even when the user may have difficulty in doing so directly due to lack ofudexperience in the query design or lack of familiarity of the search domain. A user survey has shown that theudalgorithmically formed query has recall coverage and precision characteristics better than those achieved byudthe experienced human web searchers.