Improved Semantic Retrieval Based on Domain Ontology

Semantic retrieval is a hot area of Web search technology constantly because of the growing information. This paper proposed a creative semantic retrieval method with Path Weight Semantic Distance (PWSD) algorithm which is improved from current model. Based on the domain ontology built by protege, PWSD considers both the hierarchy and the property restrictions of the concepts synthetically, and provides an accurate computation of semantic distance between conceptions. We have investigated the feasibilities, advantages and statistical significances from the comparison between our system and personalized categorization system (PCAT) and found that our system has high performances with 95% or higher statistical significance for short queries tasks.