Multi-Level Graph Attention Network Contrastive Learning for Knowledge-Aware Recommendation

2026-05-08

AI Summary

This work addresses the challenges of label sparsity, insufficient graph structure learning, and the adverse impact of noisy entities in knowledge graph-based recommendation. To tackle these issues, the authors propose a multi-view graph contrastive learning framework that integrates knowledge graph distillation to enhance user representations and employs a graph attention network to aggregate neighborhood information for generating high-quality item embeddings. The framework further introduces a three-pronged self-supervised contrastive learning mechanism operating across inter-layer, intra-layer, and interaction-level views, thereby improving intra-class generalization and inter-class discriminability. Extensive experiments on three public datasets demonstrate that the proposed method significantly outperforms state-of-the-art baselines, and ablation studies confirm the effectiveness of each component.

Abstract

In recent years, the use of edge information provided by knowledge graphs together with the advantages of higher-order connectivity in graph neural networks for recommendation systems has become an important research direction. However, existing approaches are often limited by sparse labels, insufficient graph structure learning, and noisy entities in the knowledge graph, which reduce recommendation accuracy. To address these limitations, we propose a multi-view graph contrastive learning framework. The proposed method enhances user representations through multi-view knowledge graph distillation, enabling more accurate modeling of user preferences over entities and relations. The network aggregates neighborhood entity information to construct informative item representations. Furthermore, we design a multi-level self-supervised contrastive learning module that performs comparisons across three perspectives: Inter-Level, Intra-Level, and Interaction-Level. This design improves the model's ability to generalize across intra-class samples while increasing discrimination between inter-class samples, thereby enabling more effective multi-dimensional feature modeling. We conduct extensive experiments on three public datasets using both baseline and ablation settings. Experimental results demonstrate that the proposed framework consistently outperforms existing state-of-the-art methods. Ablation studies further verify the effectiveness of each module in the proposed model.

Problem

• knowledge graph
• recommendation
• sparse labels
• noisy entities
• graph structure learning

Innovation

• multi-view contrastive learning
• knowledge graph distillation
• multi-level self-supervised learning
• graph attention network
• knowledge-aware recommendation