Hybrid Approach for Query Expansion using Query Log

Web search users usually submit short and ambiguous queries to specify their requirement. In order to improve performance of short and ambiguous queries, query expansion is used. Query expansion is as an effective way to improve the performance of information retrieval systems by adding relevant terms to the original query. After using search engine lots of data get accumulated, from which queries that have been used to retrieve documents are used. This data is stored as query log. These query logs provide valuable information to extract relationships between queries and documents that can be used in query expansion. This paper proposes method first to determine ambiguous queries using Kullback leibler distance model. It measures difference between two probability distributions. Second, relevant or most suitable expansion terms are selected from the documents with the analysis of relation between queries and documents. The relation can be evaluated by calculating frequency co-efficient with respect to document and document collection.