Clustering Using Association Word Knowledge Base and Bayesian-CUG Algorithm for Collaborative Filtering

초록

Grouping user into clusters based on the web documents they have retrieved/fetched allows accurate recomendations of new web documents. A variety of algorithms have previously been reported in the literature and their promising performance has been evaluated empirically. We identify the shortcomings of current algorithms for grouping user and propose the use of Bayesian-CUG algorithm designed in this paper. Bayesian_CUG algorithm constructs association word knowledge base that is weighted by Naive Bays learning to improve accuracy. Then, web documents retrieved/fetched by user are classified into classes by Naive Bays classifier and CUG algorithm groups users into clusters based on these web documents. As the result, Bayesian-CUG algorithm can be used to improve the efficiency of information retrieval by prefetching documents for the users and storing them in a document database in the system. We evaluate our algorithm on database of user ratings for special computer study and show that it significantly outperforms previously proposed algorithms.