A Survey of Graph Neural Networks for Social Recommender Systems

• DOI: 10.48550/arxiv.2212.04481
• Publication Date: 2022-01-01
• Authors: Kartik Sharma; Yeon-Chang Lee; Sivagami Nambi; Aditya Salian; Shlok Shah; Sang‐Wook Kim; Srijan Kumar
• Keywords: Leverage (statistics); Benchmark (surveying); Social graph

Social recommender systems (SocialRS) simultaneously leverage the user-to-item interactions as well user-to-user social relations for task of generating item recommendations to users. Additionally exploiting is clearly effective in understanding users' tastes due effects homophily and influence. For this reason, SocialRS has increasingly attracted attention. In particular, with advance graph neural networks (GNN), many GNN-based methods have been developed recently. Therefore, we conduct a comprehensive systematic review literature on SocialRS. survey, first identify 84 papers after annotating 2151 by following PRISMA framework (preferred reporting items reviews meta-analyses). Then, comprehensively them terms their inputs architectures propose novel taxonomy: (1) input taxonomy includes 5 groups type notations 7 representation notations; (2) architecture 8 GNN encoder notations, 2 decoder 12 loss function notations. We classify into several categories per describe details. Furthermore, summarize benchmark datasets metrics widely used evaluate methods. Finally, conclude survey presenting some future research directions. GitHub repository curated list are available at https://github.com/claws-lab/awesome-GNN-social-recsys.