Democratic Recommendation with User and Item Representatives Produced by Graph Condensation

2025-11-22

AI Summary

To address high computational overhead, inefficient higher-order information propagation, and structural information loss in large-scale user-item interaction graphs, this paper proposes DemoRec, a democratized graph compression framework for recommendation. Methodologically, DemoRec is the first to introduce graph compression into recommender systems: it constructs a compact representative graph via node clustering and low-order neighborhood aggregation, preserving the essential bipartite structure while reducing reliance on higher-order neighbors. Inspired by democratic mechanisms, it enables collaborative learning of representative nodes for both users and items, enhancing generalization. The framework achieves strong scalability and robustness. Extensive experiments on four public benchmarks demonstrate that DemoRec significantly outperforms state-of-the-art methods, delivering consistent improvements in recommendation accuracy, training efficiency, and robustness under data sparsity.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience
• Economics, Online Markets and Human Computation: Economics and fairness of platforms and recommendation systems

Abstract

The challenges associated with large-scale user-item interaction graphs have attracted increasing attention in graph-based recommendation systems, primarily due to computational inefficiencies and inadequate information propagation. Existing methods provide partial solutions but suffer from notable limitations: model-centric approaches, such as sampling and aggregation, often struggle with generalization, while data-centric techniques, including graph sparsification and coarsening, lead to information loss and ineffective handling of bipartite graph structures. Recent advances in graph condensation offer a promising direction by reducing graph size while preserving essential information, presenting a novel approach to mitigating these challenges. Inspired by the principles of democracy, we propose extbf{DemoRec}, a framework that leverages graph condensation to generate user and item representatives for recommendation tasks. By constructing a compact interaction graph and clustering nodes with shared characteristics from the original graph, DemoRec significantly reduces graph size and computational complexity. Furthermore, it mitigates the over-reliance on high-order information, a critical challenge in large-scale bipartite graphs. Extensive experiments conducted on four public datasets demonstrate the effectiveness of DemoRec, showcasing substantial improvements in recommendation performance, computational efficiency, and robustness compared to SOTA methods.

Problem

• Addresses computational inefficiency in large-scale user-item interaction graphs
• Mitigates information loss in bipartite graph recommendation systems
• Reduces over-reliance on high-order information in graph-based recommendations

Innovation

• Graph condensation reduces graph size and complexity
• Clustering nodes with shared characteristics for representatives
• Mitigates over-reliance on high-order information in graphs