CSCF: Clustering based-approach for social collaborative filtering

Nowadays, Collaborative Filtering (CF) has become a widely used technique in the field of recommender systems. It aims to recommend items that are relevant to the tastes and preferences of the users based on the social relationships between them. One crucial issue in CF is the Cold Start Recommendation which includes two key aspects: new user and new item. Cold user is a new comer who enters the system and cannot get relevant items, while cold item is a new item that cannot be recommended since it has no ratings yet. In this paper, we present “CSCF” a graph-based approach for social collaborative filtering. CSCF offers many interactive tasks aiming to improve the user satisfaction and solves the cold start challenges by identifying the most effective delegates with clustering. Computational results are demonstrated to confirm the effectiveness of our proposed approach.