Personalized Ranking on Cascading Behavior Graphs for Accurate Multi-Behavior Recommendation

2025-02-17

AI Summary

Existing multi-behavior recommendation methods suffer from over-smoothing and frequency bias in user behavioral representation learning (e.g., view → add-to-cart → purchase), and graph ranking techniques are ill-suited to the hierarchical, multi-relational structure of such behaviors. This paper pioneers the integration of graph ranking into multi-behavior recommendation, proposing an iterative Graph Ranking algorithm grounded in Cascaded Behavior Graph modeling and three-objective co-optimization—jointly enforcing smoothness regularization, query fidelity, and cascade alignment. The method is theoretically guaranteed to converge and achieves linear time complexity. Extensive experiments on three real-world datasets demonstrate significant improvements: +9.56% in HR@10 and +7.16% in NDCG@10 over state-of-the-art baselines. Moreover, the approach offers strong interpretability—via explicit behavioral cascade modeling—and high scalability—due to its efficient graph propagation design.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Data Mining & Knowledge Management: Recommender Systems
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Personalized content ranking and presentation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Multi-behavior recommendation predicts items a user may purchase by analyzing diverse behaviors like viewing, adding to a cart, and purchasing. Existing methods fall into two categories: representation learning and graph ranking. Representation learning generates user and item embeddings to capture latent interaction patterns, leveraging multi-behavior properties for better generalization. However, these methods often suffer from over-smoothing and bias toward frequent interactions, limiting their expressiveness. Graph ranking methods, on the other hand, directly compute personalized ranking scores, capturing user preferences more effectively. Despite their potential, graph ranking approaches have been primarily explored in single-behavior settings and remain underutilized for multi-behavior recommendation. In this paper, we propose CascadingRank, a novel graph ranking method for multi-behavior recommendation. It models the natural sequence of user behaviors (e.g., viewing, adding to cart, and purchasing) through a cascading behavior graph. An iterative algorithm computes ranking scores, ensuring smoothness, query fitting, and cascading alignment. Experiments on three real-world datasets demonstrate that CascadingRank outperforms state-of-the-art methods, with up to 9.56% and 7.16% improvements in HR@10 and NDCG@10, respectively. Furthermore, we provide theoretical analysis highlighting its effectiveness, convergence, and scalability, showcasing the advantages of graph ranking in multi-behavior recommendation.

Problem

• Multi-behavior recommendation prediction
• Overcoming over-smoothing in representation learning
• Enhancing graph ranking for multi-behavior settings

Innovation

• Cascading behavior graph modeling
• Iterative ranking score algorithm
• Multi-behavior recommendation optimization