A new approach to query expansion

摘要: Searching online text collections can be both rewarding and frustrating. On the same time valuable information found, typically many irrelevant documents are also retrieved relevant ones missed. Word mismatches between user's query document contents a main cause of retrieval failures. Expanding with related words improve search performance, but finding using is an open problem. basis previous approaches to expansion, this paper proposes new approach which combines two popular traditional methods -thesauri automatic relevance feedback. In terms theoretical analysis experiments, effective expansion for Web out-performs optimized, conventional approaches.