Research On Qurery Expansion Algorithm In Mobile Search

With the rapid development of networks and communication technologies, mobile search become more and more popular for its convenience and personality in recent years. How to get the needed information form this huge information has become a very important question. Because of the mismatch of the user's query words and the document's keywords, the result of the traditional retrieval is not satisfactory. So the research of the query expansion in order to resolve the problem of word mismatch"in mobile search is worthwhile both on theory and practice.Based on extensive and deep review of literature, a thorough analysis and resarch on theretical and application oriented problem is presetend. The material contents of the study are as follows.Firstly, this paper presents the development of mobile search. The basic theory and methods of query optimization technology are systematically and thoroughly introduced.Secondly, according to mobile search features and improving the local co-occurrence of the word query expansion algorithm, a mobile query expansion method based on word co-occurrence of the local context analysis is proposed. The method analyze the related words co-occurrence information in between the abstract of documents and query logs to evaluate quality of the expansion words and select most appropriate expansion terms.Then the expand query select the optimal weights and become the advanced query.Thirdly, on the basis of the characteristics of mobile devices, using of mobile user context-aware information to extend the user's query, a new algorithm about query expansion based on context-aware for mobile is presented to solve the problem of mobile users can not achieve accurate information in a specific environment. This algorithm based on geographically located and the objects around user to expand the query, and achieve more accurate query results.At last, for proving the effectiveness of the search algorithm, a simulation experiments is presented in this paper. Experiment results indicate that algorithm has better performance compared with other homogeneous strategies.