Sementic Retrieval Research Based on Ontology

The rapid growth and diversities of Web information bring a lot of difficulties to the efficient information-retrieval. The current information retrieval tools just offers keywords-based searching, but ignores the semantic content of the keywords itself. The author's library information retrieval system takes advantage of ontology, it expands the requirement of users to the sementic words sets and provides the document analyzer that can filter the Web pages returned by the search agent according to the certain algorithm.Consequently it presents the most relavant documents to the users.