Abstract: Query expansion is an information retrieval technique in which new query terms are added to the original query terms to improve search performance. Contextual query expansion is major issue in todayís era. In this paper, contextualization is achieved by performing document extraction and terms extraction activities to the particular domain information source. User query is expanded using document extraction and terms extraction activities. Document extraction is achieved by BM25 retrieval function. It ranks set of documents based on query terms appearing in each documents .Now second function is terms extraction process, in this, terms in the top returned documents are weighted using sub linear terms frequency scaling formula which is used to weight the terms in the expanded query derived from original query which will cope with the term mismatch problem in specific domain. Hence, this paper presents a proposal to make web searches adaptive to the context of the userís query, thus improving query results. The proposed approach makes the context acquisition faster, In addition the results of the query engine with and without the contextual information showed improvements in the precision and search length of the web results.
Keywords: BM25 ranking model, Context, Information retrieval, Internet, Query Expansion, Sub linear term frequency scaling function.