An Improved Collaborative Filtering Recommendation Method with User Trust-Preference Fusion

Collaborative filtering (CF) is the most popular recommendation technique but still suffers from data sparisity and coldstart problems.Shambour proposed a trust-semantic fused (TSF) hybrid recommender approach, which incorporated additional information from the users'social trust network and the items'semantic domain knowledge to alleviate these problems, but it involves large computation.In this paper we improved the user-based trust enhanced CF algorithm therein.By introducing a weighting combination parameter we merge the user trust weighted rating and the user preference weighted rating together to obtain the overall rating prediction.Simulation results under the Movielens datasets show the proposed method is superior to the baseline algorithms in terms of mean absolute error (MAE).