Multi-Behavior Recommender Systems: A Survey

2025-03-10

AI Summary

Traditional recommender systems often rely on a single interaction type (e.g., ratings or purchases), failing to capture users’ complex, multi-faceted interests. This survey systematically reviews Multi-Behavior Recommendation (MBR), focusing on the joint modeling of heterogeneous user behaviors—such as clicks, add-to-carts, and purchases. We propose the first unified three-stage taxonomy—Modeling, Encoding, and Training—to structurally categorize MBR methods. We comparatively analyze mainstream techniques, including graph neural networks, multi-task learning, behavioral sequence modeling, meta-path-guided embedding, and contrastive learning, identifying cross-behavior semantic alignment and dynamic behavior weighting as key challenges. Empirical analysis of over 100 studies demonstrates that multi-behavior collaboration consistently improves core metrics—including click-through rate (CTR) and conversion rate. This work establishes a structured conceptual framework for MBR research and delivers a reusable, industry-ready methodology guide for building robust, behavior-aware recommender systems.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

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

Abstract

Traditional recommender systems primarily rely on a single type of user-item interaction, such as item purchases or ratings, to predict user preferences. However, in real-world scenarios, users engage in a variety of behaviors, such as clicking on items or adding them to carts, offering richer insights into their interests. Multi-behavior recommender systems leverage these diverse interactions to enhance recommendation quality, and research on this topic has grown rapidly in recent years. This survey provides a timely review of multi-behavior recommender systems, focusing on three key steps: (1) Data Modeling: representing multi-behaviors at the input level, (2) Encoding: transforming these inputs into vector representations (i.e., embeddings), and (3) Training: optimizing machine-learning models. We systematically categorize existing multi-behavior recommender systems based on the commonalities and differences in their approaches across the above steps. Additionally, we discuss promising future directions for advancing multi-behavior recommender systems.

Problem

• Enhance recommendation quality using diverse user-item interactions.
• Review data modeling, encoding, and training in multi-behavior systems.
• Categorize and discuss future directions for multi-behavior recommenders.

Innovation

• Leverages diverse user-item interactions
• Transforms multi-behaviors into embeddings
• Optimizes machine-learning models systematically