Disentangled Modeling of Social Homophily and Influence for Social Recommendation
Nian Li, Chen Gao, Depeng Jin, Qingmin Liao
Abstract
Social recommendation leverages social information to alleviate data sparsity and cold-start issues of collaborative filtering (CF) methods. Most existing works model user interests following the assumption of <i>social homophily</i> based on social-relation data. The explicit modeling of <i>social influence</i>, which also largely affects user behaviors, has not been well explored. Considering user behaviors may be driven by social factors in today’s information services (<i>e.g.</i>, purchasing products shared by close friends on social e-commerce applications), these methods will be suboptimal. In this work, we propose a method modeling both social homophily-aware user interests and social influence as two essential effects on user behaviors for social recommendation, named as DISGCN (short for <b>DIS</b>entangled modeling of Social homophily and influence with <b>G</b>raph <b>C</b>onvolutional <b>N</b>etwork). Specifically, we devise a disentangled embedding layer to encode these two effects. Furthermore, two tailored graph convolutional layers are developed to disentangle them refinedly, leveraging the high-order embedding propagation in social-network graph from two aspects. Technically, first, the operation of attentive embedding propagation is adopted for capturing personalized social homophily-aware interests, and second, the item-gate-based embedding propagation is proposed for capturing item-specific social influence. In addition, to ensure the disentanglement of social influence, we propose a contrastive learning framework that endows corresponding embeddings with explicit semantics. Extensive experiments on two real-world datasets demonstrate the effectiveness of our proposed model. Further studies also verify the rationality and necessity of our designs. We have released the datasets and codes at this link: <uri>https://github.com/tsinghua-fib-lab/DISGCN</uri>.