Abstract:A well-developed recommendation system can not only leverage multi-typed interactions (such as page view, add-to-cart, and purchase) to better identify user preferences, but also demonstrate high performance, low complexity, and strong interpretability. However, many existing solutions for multi-behavior recommendation fall short of intuitive modeling of real-world scenarios, leading to overly complex models with massive parameters and cumbersome components. In particular, they share two critical limitations: (1) Some pioneering models are built upon the strict assumption of cascade effects across behaviors, which contradicts multifarious behavior paths in practical applications. (2) Existing approaches fail to explicitly capture the unique idiosyncrasies of users and even neglect the inherent nature of items involved in the multi-behavior interactions. To this end, we propose a novel Directed Acyclic Graph Convolutional Network (DA-GCN) for the multi-behavior recommendation task. Specifically, we pinpoint the partial order relations within the monotonic behavior chain and extend it to personalized directed acyclic behavior graphs to exploit behavior dependencies. Then, a GCN-based directed edge encoder is employed to distill rich collaborative signals embodied by each directed edge. In light of the information flows over the directed acyclic structure, we propose an attentive aggregation module to gather messages from all potential antecedent behaviors, representing distinct perspectives to understand the terminated behavior. Thus, we obtain comprehensive representations for the follow-up behavior through learnable distributions over its preceding behaviors, explicitly reflecting personalized interactive patterns of users and underlying properties of items simultaneously. Finally, we design a customized multi-task learning objective for flexible joint optimization. Extensive experiments on public benchmarking datasets fully demonstrate the superiority of DA-GCN with significant performance improvement and computational efficiency over a wide range of state-of-the-art methods. Our code is available at https://github.com/xizhu1022/DA-GCN.

Global Heterogeneous Graph and Target Interest Denoising for Multi-behavior Sequential Recommendation.

Multi-behavior Graph Contextual Aware Network for Session-based Recommendation

Dual Graph Attention Networks for Multi-Behavior Recommendation

Global and Personalized Graphs for Heterogeneous Sequential Recommendation by Learning Behavior Transitions and User Intentions

Denoising and Prompt-Tuning for Multi-Behavior Recommendation

MB-HGCN: A Hierarchical Graph Convolutional Network for Multi-behavior Recommendation

HEC-GCN: Hypergraph Enhanced Cascading Graph Convolution Network for Multi-Behavior Recommendation

Exploiting Behavior Multiplicity for Social Recommendation Via Graph Neural Networks

Self-supervised progressive graph neural network for enhanced multi-behavior recommendation

END4Rec: Efficient Noise-Decoupling for Multi-Behavior Sequential Recommendation

Research on Efficient Multi-Behavior Recommendation Method Fused with Graph Neural Network

Multi-Behavior Dynamic Contrastive Learning for Recommendation

Multipath-guided Heterogeneous Graph Neural Networks for Sequential Recommendation

Multi-Grained Fusion Graph Neural Networks for Sequential Recommendation.

A GCN-Based Sequential Recommendation Model for Heterogeneous Information Networks

Intention Enhanced Dual Heterogeneous Graph Attention Network for Sequential Recommendation

Cascading Residual Graph Convolutional Network for Multi-Behavior Recommendation

Hybrid Graph Neural Network Recommendation Based on Multi-Behavior Interaction and Time Sequence Awareness

Multi-Behavior Recommendation with Personalized Directed Acyclic Behavior Graphs

Multi-behavior Enhanced Graph Neural Networks for Social Recommendation.

Novel Behavior-Enhanced Long- and Short-Term Interest Model for Sequential Recommendation