Query Expansion Based On Both Long-Term And Short-Term Social Relation In XML Social Book Search

Query expansion technology is usually used to make abundant query to search information. Traditional query expansion based on WordNet, user search log or user feedback couldn’t get the user’s real intent, sometimes the latter could expand overly. In this paper, we propose a novel query expansion method based both on the user’s long-term relation and short-term relation in XML social book search. We first propose recommendation degree of a tagged word based on a user’s long-term social relation and short-term relation respectively, and the combined recommendation degree of a tagged word, then based on these definitions, we propose a new query expansion algorithm. The experimental results on real-world data show that our method has greatly increased the retrieval precise and users’ satisfaction . Keywordsquery expansion; long-term social relation; short-term social relation; book search; recommendation degree