An Improved Collaborative Filtering Combining User Preference and Trust
Shukun Niu · 2018
To solve the sparseness problem which affects the performance of the recommendation system, a collaborative filtering algorithm combining user preference and trust degree is proposed. The rating grade, rating authority and rating recognition are used to calculate the trust between users which integrates with Pearson similarity by weighting to obtain the trusted similarity. In addition, the information entropy theory is introduced to obtain the difference information entropy between users according to the user-item rating matrix. At last, the trusted similarity and the difference information entropy are weighted to obtain the overall similarity which used to predict the rating of the target user. The simulation shows that the proposed algorithm has a higher performance than the traditional collaborative filtering algorithm which measures the similarity between users using PCC.