HGCL: Hierarchical Graph Contrastive Learning for User-Item Recommendation

2025-05-25

AI Summary

Existing graph contrastive learning (GCL) methods for user–item recommendation overlook the multi-scale hierarchical structure of items, limiting their ability to model similarity relationships across fine-grained to coarse-grained semantic levels. To address this, we propose the first GCL framework that explicitly incorporates a two-level item hierarchy: a dual-layer bipartite graph is constructed via representation compression and hierarchical clustering; then, cross-layer contrastive pretraining and hierarchical graph fine-tuning are jointly optimized to enhance both discriminability and semantic consistency of learned representations. Our approach explicitly encodes item-level structural semantics into the contrastive objective. Extensive experiments on three large-scale benchmark datasets (70K–382K nodes) demonstrate significant improvements over state-of-the-art methods, validating that explicit modeling of item hierarchy yields critical gains in recommendation accuracy and generalization.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Data Mining & Knowledge Management: Recommender Systems

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Graph Contrastive Learning (GCL), which fuses graph neural networks with contrastive learning, has evolved as a pivotal tool in user-item recommendations. While promising, existing GCL methods often lack explicit modeling of hierarchical item structures, which represent item similarities across varying resolutions. Such hierarchical item structures are ubiquitous in various items (e.g., online products and local businesses), and reflect their inherent organizational properties that serve as critical signals for enhancing recommendation accuracy. In this paper, we propose Hierarchical Graph Contrastive Learning (HGCL), a novel GCL method that incorporates hierarchical item structures for user-item recommendations. First, HGCL pre-trains a GCL module using cross-layer contrastive learning to obtain user and item representations. Second, HGCL employs a representation compression and clustering method to construct a two-hierarchy user-item bipartite graph. Ultimately, HGCL fine-tunes user and item representations by learning on the hierarchical graph, and then provides recommendations based on user-item interaction scores. Experiments on three widely adopted benchmark datasets ranging from 70K to 382K nodes confirm the superior performance of HGCL over existing baseline models, highlighting the contribution of hierarchical item structures in enhancing GCL methods for recommendation tasks.

Problem

• Incorporates hierarchical item structures for better recommendations
• Enhances GCL with cross-layer contrastive learning
• Improves accuracy by modeling multi-resolution item similarities

Innovation

• Hierarchical item structures enhance GCL
• Cross-layer contrastive learning pre-trains representations
• Two-hierarchy graph fine-tunes recommendation scores