A New Collaborative Filtering Recommendation Approach Based on Naive Bayesian Method Kebin Wang and Ying Tan Key Laboratory of Machine Perception (MOE), Peking University Department of Machine Intelligence, School of Electronics Engineering and Computer Science, Peking University, Beijing, 100871, China caesar1017@gmail.com, ytan@pku.edu.cn Abstract. Recommendation is a popular and hot problem in e-commerce. Recommendation systems are realized in many ways such as content-based recommendation, collaborative filtering recommendation, and hybrid approach recommendation. In this article, a new collaborative filtering recommendation algorithm based on naive Bayesian method is proposed. Unlike original naive Bayesian method, the new algorithm can be applied to instances where conditional independence assumption is not obeyed strictly. According to our experiment, the new recommendation algorithm has a better performance than many existing algorithms including the popular k-NN algorithm used by Amazon.com especially at long length recommendation. Keywords: recommender system, collaborative filtering, naive Bayesian method, probability. 1 Introduction Recommendation systems are widely used by e-commerce web sites. They are a kind of information retrieval. But unlike search engines or databases they pro- vide users with things they have never heard of before. That is, recommendation systems are able to predict users’ unknown interests according to their known interests[8],[10]. There are thousands of movies that are liked by millions of peo- ple. Recommendation systems are ready to tell you which movie is of your type out of all these good movies. Though recommendation systems are very useful, the current systems still require further improvement. They always provide ei- ther only most popular items or strange items which are not to users’ taste at all. Good recommendation systems have a more accurate prediction and lower computation complexity. Our work is mainly on the improvement of accuracy. Naive Bayesian method is a famous classification algorithm[6] and it could also be used in the recommendation field. When factors affecting the classification results are conditional independent, naive Bayesian method is proved to be the solution with the best performance. When it comes to the recommendation field, naive Bayesian method is able to directly calculate the probability of user’s possible interests and no definition of similarity or distance is required, while in Y. Tan et al. (Eds.): ICSI 2011, Part II, LNCS 6729, pp. 218–227, 2011. c  Springer-Verlag Berlin Heidelberg 2011