PHKT:Personalized Dynamic Hypergraph-enhanced KAN-Transformer for Multi-behavior Sequential Recommendation

2026-06-03

AI Summary

This work addresses the limitations in existing multi-behavior sequential recommendation methods, particularly insufficient modeling of heterogeneous semantics, lack of user-specific weighting, and inadequate capture of sequential dependencies. To overcome these challenges, the authors propose PHKT, a novel model featuring a personalized dynamic hypergraph module that adaptively captures user-specific high-order relationships. Furthermore, PHKT is the first to integrate Kolmogorov–Arnold Networks (KANs) into the feed-forward network of a Transformer architecture, thereby enhancing fine-grained nonlinear representation of latent patterns across heterogeneous behaviors. Extensive experiments on three real-world datasets—Tmall, RetailRocket, and IJCAI—demonstrate that PHKT significantly outperforms nine strong baseline models across multiple evaluation metrics, confirming its effectiveness in modeling multi-behavior user preferences and predicting target actions.

Abstract

In multi-behavior recommendation, auxiliary behaviors such as clicks, add-to-cart, and purchases can provide richer supervisory information for predicting target behaviors. Although existing graph and hypergraph methods are capable of modeling high-order relationships among users, items, and behaviors, they still have limitations in heterogeneous semantics, user-specific weighting, and sequence dependency modeling. While standard Transformers excel at sequence modeling, their shared feedforward mapping struggles to accommodate the differentiated requirements of heterogeneous latent patterns in multi-behavior scenarios. To address this, this paper proposes the Personalized Hypergraph-enhanced Kolmogorov-Arnold Network Transformer (PHKT). Specifically, we design a personalized dynamic hypergraph module that performs behavior-aware weighting of item similarities based on users' historical behavior sequences to capture user-specific heterogeneous high-order relationships. Meanwhile, a Transformer is used as the temporal backbone to model the evolution of short- and long-term preferences, and KAN is introduced to replace the traditional MLP in the feedforward network to enhance fine-grained modeling capability for nonlinear responses to different latent patterns. Experiments on three real datasets, Tmall, RetailRocket, and IJCAI, show that PHKT consistently outperforms nine strong baseline models across multiple evaluation metrics, demonstrating its effectiveness in multi-behavior preference modeling and target behavior prediction.

Problem

• multi-behavior recommendation
• heterogeneous semantics
• user-specific weighting
• sequence dependency modeling
• high-order relationships

Innovation

• Personalized Hypergraph
• Kolmogorov-Arnold Network (KAN)
• Multi-behavior Recommendation
• Transformer
• Dynamic Item Weighting