Abstract:

The personalized recommendation technique addresses studying the behaviors of individual users, analyzing what they are interested in and recommending suitable resources to them. In other words, the personalized recommendation technique is a better solution to the contradiction between the requirements of users and the explosive information on the Internet. Collaborative filtering algorithms based on neighborhood approach and potential factors are the most successful techniques in the recommendation system. Although the former is easy to implement, its accuracy needs to be improved. Meanwhile, the latter has high precision, but it is complex and the parameters are difficult to learn. Therefore, in the paper, an improved collaborative filtering algorithm based on user similarity is proposed. Through adjusting the measure method of user similarity, it can generate more reasonable user neighbors and recommend the users according to their scores. Experimental results show that the algorithm proposed in this paper is easier to implement than the algorithm based on the potential factors. Besides, compared with the algorithm based on the neighborhood approach, our proposal, to some extent, improves the accuracy.

Key words: recommendation system;collaborative filtering;user similarity;user neighbor