The Research On Tagging Recommendation System Based On Semantic

With the rapid development of the Internet and Web2.0 technology, the problem of information overloading and information amazing, which we are having been frustrated, has became worse. And all these boost the flourishing development of personalized recommendation system. Present personalized recommendation technology has eased the pressure and cut down cost for people finding their interested information. But the tradition recommender system couldn't really understand our users' preference in certain extend. It may cause the inaccuracy of result of the recommendation.This paper considers the limitations of the tradition recommender system; we combine the tagging system and recommender system, and propose a framework of tagging recommender system. We analyze the model of our framework, from the structure of tags, relevant tags selecting algorithm, to construct user preference model based on tags. And in the part of recommender algorithm, we propose a similarity computing based on semantic web. We also solve the problem about similarity between words and similarity between sentences.In the end, we validate the validity of our proposed algorithms through experiments, and compare with present recommendation algorithms. The results show that our methods outperform the existing methods in recommendation precise and time cost, which contribute the excellent performance of personalized recommendation system.