A Novel Algorithm for Group Recommendation Based on Combination of Recessive Characteristics

Recommendation system is an effective way to solve the problem of information overload, and it was widely used in the real life. In view of user groups have complex characteristics, many studies found that some important recessive characteristics could more truly reflect the interests of group users. In this paper, we propose a novel algorithm for group recommendation based on combination of recessive characteristics (GRBCRC). Firstly, on the basis of SVD++ recommendation algorithm, we combine the recessive characteristics and propose a novel personal recommendation (NPR) algorithm as the basis of group recommendation algorithm. By the NPR algorithm, we get the user's prediction scores, and then we propose GRBCRC by combining the group weight calculation based on recessive characteristics. The experimental results show the better performance of our proposed algorithm.