Domain Ontology-based Semantic Retrieval And Personalized Recommendation Algorithm

This dissertation analyzes the current information retrieval technology,for the problem of the information retrieval system which based on keyword matching does not guarantee high-quality output, through the used of Semantic Web and ontology technology to build the semantic retrieval ontology-based system, raise the level from concept to the semantic level, thereby eliminating the semantic gap between the needs of the user retrieve and the computer, achieve an accurate understanding of user needs, return a high-quality search results to the user. And then based on the semantic retrieval system,try to use ontology to describe the film characteristics, by the calculating of the similarity of the film characteristics, we can predict the user's interest and take personalized recommendations to the user.Firstly, base on the research of the Semantic Web and Ontology, use the Web Ontology Language OWL construct a domain ontology, and then introduced the design of the ontology classes, properties and individual in detail.Secondly, base on the builded Ontology, introduced the design of a semantic retrieval system framework,and the implement of each module oth the retrieval system, including resource acquisition module, ontology analysis module, ontology reasoning and query modules. Through the function testing of the retrieval system, the semantic search functions are verified.Finally, based on the need of film retrieval system, proposed an ontology-based personalized recommendation algorithm, introduce the realization of the algorithms for a in detail, through the analysis and evaluate of the computational results, the applicability of the algorithm is verified.