Recommender Systems with Characterized Social Regularization

TL;DR

This paper introduces CSR, a social recommendation model that accounts for variable social influence, improving recommendation accuracy by modeling diverse user-friend preferences.

Contribution

The paper proposes a novel regularization approach that captures the heterogeneity of social influence in recommendation systems, applicable to both explicit and implicit feedback.

Findings

CSR outperforms existing methods on real-world datasets.

The model effectively captures diverse social influences.

Significant improvement in recommendation accuracy.

Abstract

Social recommendation, which utilizes social relations to enhance recommender systems, has been gaining increasing attention recently with the rapid development of online social network. Existing social recommendation methods are based on the fact that users preference or decision is influenced by their social friends' behaviors. However, they assume that the influences of social relation are always the same, which violates the fact that users are likely to share preference on diverse products with different friends. In this paper, we present a novel CSR (short for Characterized Social Regularization) model by designing a universal regularization term for modeling variable social influence. Our proposed model can be applied to both explicit and implicit iteration. Extensive experiments on a real-world dataset demonstrate that CSR significantly outperforms state-of-the-art social recommendation methods.