An Improved Trust-aware Recommender System for Personalized User Recommendation in Tmall

Abstract

Annual Double Eleven shopping festival has attracted people's attention greatly. With the increase of online shopping channel, trust has become the most important content of user interaction in this environment. In this paper, We proposed improved trust-aware recommender system (iTARS) produces valuable recommendations by dynamic trust between users and selecting a best neighborhood based on biological metaphor of ant colonies in Tmall. The performance of iTARS is evaluated using tmall datasets of differernt sparsity levels and compared with traditional trust-aware recommender system for generating recommendations to the tmall members.