A trust-aware group recommender system using particle swarm optimization

With the exponential growth of the online community activities, group recommender systems have become popular in recent years. However, making recommendations relevant to the common interests of a group is a challenging task due to the diversity of group members' preferences. In this paper, we propose a novel Trust-aware Group Recommendation (TGR) approach to improve the performance of group recommendations. TGR uses a new group trust metric that is optimized by Particle Swarm Optimization (PSO). This metric directly provides a set of neighbors for a group of users. The experimental results show that TGR can improve the accuracy and run-time performance of other group recommender systems.