1 code implementation • 26 Aug 2022 • Shaowen Peng, Kazunari Sugiyama, Tsunenori Mine
With the tremendous success of Graph Convolutional Networks (GCNs), they have been widely applied to recommender systems and have shown promising performance.
1 code implementation • 24 Apr 2022 • Shaowen Peng, Kazunari Sugiyama, Tsunenori Mine
To unveil the effectiveness of GCNs for recommendation, we first analyze them in a spectral perspective and discover two important findings: (1) only a small portion of spectral graph features that emphasize the neighborhood smoothness and difference contribute to the recommendation accuracy, whereas most graph information can be considered as noise that even reduces the performance, and (2) repetition of the neighborhood aggregation emphasizes smoothed features and filters out noise information in an ineffective way.
no code implementations • 30 Apr 2020 • Shaowen Peng, Tsunenori Mine
Graph Convolutional Network (GCN) has achieved great success and has been applied in various fields including recommender systems.