HPCF: Hybrid Music Group Recommendation System based on Item Popularity and Collaborative Filtering

Group recommendation system is designed to pleasure a certain group of people by providing recommendation lists to prevent information overload. This paper proposes HPCF, which combines item popularity and preference of the group members to optimize the quality of recommendation. Evaluation results show that the HPCF outperforms popularity-based methods and Singular Value Decomposition based methods. In our experiments, HPCF scheme improves the accuracy of recommendation by 8.7% to 19.8% in different conditions.