User interest driven semantic query expansion for effective web search

Retrieving user-relevant content from a large volume of data available on the Web via an input query is a difficult task. A user query may not be able to specify user information needs due to the ambiguous and limited number of query terms. The semantic query expansion (QE) strategy offers a solution to this problem by expanding the query with additional terms, which are semantically similar to the original query. However, this strategy does not consider individual user interest in the generation of expansion terms. In this article, semantic QE is improved by combining the notion of ontology knowledge and user interest. The proposed semantic QE technique involves a computing domain of the input query via ontology, generates expansion terms from the user browsing history, and finally selects expansion terms that represent user preferences on the basis of the semantic similarity between expansion terms and query and user feedback. The experimental evaluation indicates that expanded queries produced by the proposed technique retrieve more personalized contents over Web search than initial user queries. The obtained results achieve 86.4% average precision, which proves a positive impact of incorporating user preferences in semantic QE