GraphFusionSBR: Denoising Multi-Channel Graphs for Session-Based Recommendation

TL;DR

GraphFusionSBR introduces a multi-channel approach combining knowledge graphs, hypergraphs, and line graphs to effectively denoise sessions and improve recommendation accuracy in session-based systems.

Contribution

It proposes a novel multi-channel model with adaptive edge removal and mutual information maximization to address noise and item dominance issues in session-based recommendation.

Findings

Enhanced recommendation accuracy across domains

Effective noise reduction in session data

Improved handling of item dominance

Abstract

Session-based recommendation systems must capture implicit user intents from sessions. However, existing models suffer from issues such as item interaction dominance and noisy sessions. We propose a multi-channel recommendation model, including a knowledge graph channel, a session hypergraph channel, and a session line graph channel, to capture information from multiple sources. Our model adaptively removes redundant edges in the knowledge graph channel to reduce noise. Knowledge graph representations cooperate with hypergraph representations for prediction to alleviate item dominance. We also generate in-session attention for denoising. Finally, we maximize mutual information between the hypergraph and line graph channels as an auxiliary task. Experiments demonstrate that our method enhances the accuracy of various recommendations, including e-commerce and multimedia recommendations. We release the code on GitHub for reproducibility.\footnote{https://github.com/hohehohe0509/DSR-HK}