User Hesitation and Negative Transfer in Multi-Behavior Recommendation

2025-11-08

AI Summary

In multi-behavior recommendation, weak signals—where users engage in auxiliary behaviors without triggering the target behavior—reflect user hesitation and negative transfer effects, yet existing methods inadequately model them. This paper proposes the Hesitation-aware Negative Transfer (HNT) framework, the first to explicitly distinguish *positive hesitation* (indicating latent interest) from *negative interference* (reflecting preference suppression) within weak signals. HNT constructs a hesitation item set and explicitly models negative transfer effects. Methodologically, it integrates multi-behavior feature learning, item similarity modeling, and auxiliary feature fusion to enable dual-dimensional collaborative modeling of positive and negative weak signals. Extensive experiments on three real-world datasets demonstrate that HNT outperforms the best baseline by 12.57% in HR@10 and 14.37% in NDCG@10, significantly improving the identification of users’ latent intentions.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Data Mining & Knowledge Management: Recommender Systems
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Web Mining and Content Analysis: Content-based information diffusion
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Multi-behavior recommendation aims to integrate users'interactions across various behavior types (e.g., view, favorite, add-to-cart, purchase) to more comprehensively characterize user preferences. However, existing methods lack in-depth modeling when dealing with interactions that generate only auxiliary behaviors without triggering the target behavior. In fact, these weak signals contain rich latent information and can be categorized into two types: (1) positive weak signals-items that have not triggered the target behavior but exhibit frequent auxiliary interactions, reflecting users'hesitation tendencies toward these items; and (2) negative weak signals-auxiliary behaviors that result from misoperations or interaction noise, which deviate from true preferences and may cause negative transfer effects. To more effectively identify and utilize these weak signals, we propose a recommendation framework focused on weak signal learning, termed HNT. Specifically, HNT models weak signal features from two dimensions: positive and negative effects. By learning the characteristics of auxiliary behaviors that lead to target behaviors, HNT identifies similar auxiliary behaviors that did not trigger the target behavior and constructs a hesitation set of related items as weak positive samples to enhance preference modeling, thereby capturing users'latent hesitation intentions. Meanwhile, during auxiliary feature fusion, HNT incorporates latent negative transfer effect modeling to distinguish and suppress interference caused by negative representations through item similarity learning. Experiments on three real-world datasets demonstrate that HNT improves HR@10 and NDCG@10 by 12.57% and 14.37%, respectively, compared to the best baseline methods.

Problem

• Modeling user hesitation from auxiliary behaviors without target interactions
• Identifying negative transfer effects caused by misoperations or noise
• Improving recommendation accuracy through weak signal learning

Innovation

• Models user hesitation through auxiliary behavior analysis
• Identifies negative transfer effects via similarity learning
• Enhances preference modeling using weak signal learning