Abstract:Collaborative filtering is one of the most important technologies in E-commerce. With the development of E-commerce, the magnitudes of users and commodities grow rapidly, the problem of data sparsity of user project is becoming more and more significant. In traditional collaborative filtering recommender systems, similarity of users is often calculated based on common ratings. When user-item ratings are sparse, the accuracy of recommendations will be influenced because users with similar preferences can't be found accurately. Considering the effect of users' ratings and trusts on the recommendation results, this paper applies AHP to construct user trust model and proposes a collaborative filtering recommendation method combining user trust model. The experimental results show that, user similarity calculation method combining user trust can effectively reflect the user's cognitive changes, ease the impact of data sparsity on the collaborative filtering recommendation algorithm and improve the accuracy of recommendation results.