A Collaborative Filtering Algorithm based on Entropy and Trust Optimized Neighbor Selection

The Collaborative filtering (CF) is an emerging technology that deals with overload problems by providing recommendations for products that users may be interested in. Neighbor selection is a key part of the recommendation process. The traditional collaborative filtering algorithm not only does not consider the recommendation contribution ability of neighbor users to the target user, but also does not consider the impact of the user’s rate difference, which leads to untrusted similarity and increases the proportion of pseudo-nearest neighbors in the neighbor set. Aiming at these problems, a collaborative filtering recommendation algorithm based on entropy and trust optimization of neighbors is proposed. The proposed method uses user entropy to calculate user rating distribution characteristics, and measures the recommendation contribution ability between users through the difference in rating distribution. Analyze the difference in rating differences between users through the method of rapid prediction, and tap the trust among users. Finally, the dual criteria are used to jointly calculate the recommended weights and construct a set of nearest neighbors. Empirical outcomes substantiate that the proposed method accurately selects the set of nearest neighbors and significantly improves recommendation accuracy of traditional collaborative filtering algorithms.