Research On Recommendation System Technology And Its Application In Digital Library

With the rapid development of Internet technology, information on the network is growing at an unprecedented speed, and people are facing the problem of "information overload". Recommendation system can help people conveniently filter out the resources from a lot of information what they are interested in. Digital library provides users with convenient services of its information resources using network. Applying recommendation system technology to the field of digital library can realize the personalized recommendation and enhance the user?s knowledge discovery capabilities and provide users of the digital library with fast and accurate knowledge updating and retrieving service, besides the digital storage and retrieval of library resources.Collaborative filtering is one of the most widely used methods in the field of personalized recommendation. Constrained probabilistic matrix factorization algorithm is a model-based collaborative filtering algorithm which can effectively deal with the problem of scalability in large-scale recommendation system and guarantee the real-time of recommendation. This thesis takes the relationship of item-neighborhood into the algorithm model of constrained probability matrix factorization to improve the quality of the proposed algorithm. To guarantee the accuracy of item-neighborhood, the inherent features extracted from the item’s summary and the tag of user marked for the item are used to get the set of the nearest neighbor for the items, then the item-neighborhood set is applied into the framework of the constrained probabilistic matrix factorization algorithm. The experiments show that the proposed method can improve the accuracy of the original algorithm.The book retrieval function is commonly used in digital library. This thesis not only considers the correlation between results and the retrieval keywords but also takes the user?s preference for the retrieved results into consideration, so as to achieve the personalized recommendation based on the retrieval results. Finally, this thesis gives the recommendation module of digital library using the recommendation algorithm model and the personalized recommendation based on the book retrieval described above.