Collaborative Filtering Meets Spectrum Shift: Connecting User-Item Interaction with Graph-Structured Side Information

TL;DR

This paper introduces Spectrum Shift Correction (SSC), a spectral method that effectively integrates graph-structured side information into collaborative filtering models by addressing spectrum shift issues, leading to improved recommendation performance.

Contribution

The paper proposes SSC, a novel spectral correction technique that adapts GNNs to spectrum shifts caused by side information, unifying various data types without extra computational costs.

Findings

Achieves up to 23% relative improvement in recommendation accuracy.

Effectively handles spectrum shifts caused by side information.

No additional computational overhead required.

Abstract

Graph Neural Networks (GNNs) have demonstrated their superiority in collaborative filtering, where the user-item (U-I) interaction bipartite graph serves as the fundamental data format. However, when graph-structured side information (e.g., multimodal similarity graphs or social networks) is integrated into the U-I bipartite graph, existing graph collaborative filtering methods fall short of achieving satisfactory performance. We quantitatively analyze this problem from a spectral perspective. Recall that a bipartite graph possesses a full spectrum within the range of [-1, 1], with the highest frequency exactly achievable at -1 and the lowest frequency at 1; however, we observe as more side information is incorporated, the highest frequency of the augmented adjacency matrix progressively shifts rightward. This spectrum shift phenomenon has caused previous approaches built for the full spectrum [-1, 1] to assign mismatched importance to different frequencies. To this end, we propose Spectrum Shift Correction (dubbed SSC), incorporating shifting and scaling factors to enable spectral GNNs to adapt to the shifted spectrum. Unlike previous paradigms of leveraging side information, which necessitate tailored designs for diverse data types, SSC directly connects traditional graph collaborative filtering with any graph-structured side information. Experiments on social and multimodal recommendation demonstrate the effectiveness of SSC, achieving relative improvements of up to 23% without incurring any additional computational overhead. Our code is available at https://github.com/yhhe2004/SSC-KDD.