Challenging Low Homophily in Social Recommendation

TL;DR

This paper identifies the low preference-aware homophily in social graphs used for recommendations and proposes SHaRe, a graph rewiring framework with contrastive learning, to enhance social recommendation models by better capturing social relations.

Contribution

It introduces a novel data-centric framework, SHaRe, that improves social recommendation by rewiring social graphs to emphasize preference-aware homophily and integrating contrastive learning for better user representations.

Findings

SHaRe improves recommendation performance across different homophily ratios.

The framework enhances existing state-of-the-art social recommendation models.

Rewiring social graphs effectively captures preference-aware relations.

Abstract

Social relations are leveraged to tackle the sparsity issue of user-item interaction data in recommendation under the assumption of social homophily. However, social recommendation paradigms predominantly focus on homophily based on user preferences. While social information can enhance recommendations, its alignment with user preferences is not guaranteed, thereby posing the risk of introducing informational redundancy. We empirically discover that social graphs in real recommendation data exhibit low preference-aware homophily, which limits the effect of social recommendation models. To comprehensively extract preference-aware homophily information latent in the social graph, we propose Social Heterophily-alleviating Rewiring (SHaRe), a data-centric framework for enhancing existing graph-based social recommendation models. We adopt Graph Rewiring technique to capture and add highly homophilic social relations, and cut low homophilic (or heterophilic) relations. To better refine the user representations from reliable social relations, we integrate a contrastive learning method into the training of SHaRe, aiming to calibrate the user representations for enhancing the result of Graph Rewiring. Experiments on real-world datasets show that the proposed framework not only exhibits enhanced performances across varying homophily ratios but also improves the performance of existing state-of-the-art (SOTA) social recommendation models.