A curated list of papers on GNN-based social recommender systems.

Social recommender systems use user-user social links in addition to user-item interaction edges to make effective recommendation. This is illustrated in the diagram shown below, as it forms a user-item bipartite graph along with a relational graph induced by the social network.

Over the past few years, Graph Neural Networks (GNNs) have emerged as the standard way to encode graphs.

This repo, which accompanies our survey on the topic, inludes over 80 papers that use Graph Neural Network for the task of social recommendation.

You may refer to the survey for novel taxonomies to categorize these papers based on inputs, input representations, architectures (both encoders and decoders), and loss functions.

@article{sharma2021task,
   title={A Survey of Graph Neural Networks for Social Recommender Systems},
   author={Sharma, Kartik and Lee, Yeon-Chang and Nambi, Sivagami and Salian, Aditya and Shah, Shlok and Kim, Sang-Wook and Kumar, Srijan},
   journal={arXiv preprint arXiv:2212.04481},
   year={2022} 
 }

Papers sorted by year: 2022 | 2021 | 2020 | 2019

[Back to top]

[Back to top]

[Back to top]

[Back to top]