DisenHAN: Disentangled Heterogeneous Graph Attention Network for Recommendation

Abstract

Heterogeneous information network has been widely used to alleviate sparsity and cold start problems in recommender systems since it can model rich context information in user-item interactions. Graph neural network is able to encode this rich context information through propagation on the graph. However, existing heterogeneous graph neural networks neglect entanglement of the latent factors stemming from different aspects. Moreover, meta paths in existing approaches are simplified as connecting paths or side information between node pairs, overlooking the rich semantic information in the paths. In this paper, we propose a novel disentangled heterogeneous graph attention network DisenHAN for top-$N$ recommendation, which learns disentangled user/item representations from different aspects in a heterogeneous information network. In particular, we use meta relations to decompose high-order connectivity between node pairs and propose a disentangled embedding propagation layer which can iteratively identify the major aspect of meta relations. Our model aggregates corresponding aspect features from each meta relation for the target user/item. With different layers of embedding propagation, DisenHAN is able to explicitly capture the collaborative filtering effect semantically. Extensive experiments on three real-world datasets show that DisenHAN consistently outperforms state-of-the-art approaches. We further demonstrate the effectiveness and interpretability of the learned disentangled representations via insightful case studies and visualization.

arXiv commentAccepted at CIKM2020

References in corpus (16)

• Semi-Supervised Classification with Graph Convolutional Networks
• Inductive Representation Learning on Large Graphs
• BPR: Bayesian Personalized Ranking from Implicit Feedback
• Neural Graph Collaborative Filtering
• KGAT: Knowledge Graph Attention Network for Recommendation
• InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
• Graph Convolutional Matrix Completion
• AutoInt: Automatic Feature Interaction Learning via Self-Attentive Neural Networks
• Unifying Knowledge Graph Learning and Recommendation: Towards a Better Understanding of User Preferences
• Learning over Knowledge-Base Embeddings for Recommendation
• Learning Heterogeneous Knowledge Base Embeddings for Explainable Recommendation
• Deep Variational Information Bottleneck
• Variational Lossy Autoencoder
• Learning Disentangled Representations for Recommendation
• Graph Capsule Convolutional Neural Networks
• Heterogeneous Graph Transformer

Cited by in corpus (16)

• A Survey on Accuracy-oriented Neural Recommendation: From Collaborative Filtering to Information-rich Recommendation
• Self-Supervised Hypergraph Transformer for Recommender Systems
• Disentangled Contrastive Collaborative Filtering
• Disentangling Long and Short-Term Interests for Recommendation
• Multi-view Intent Disentangle Graph Networks for Bundle Recommendation
• DisenPOI: Disentangling Sequential and Geographical Influence for Point-of-Interest Recommendation
• DisenKGAT: Knowledge Graph Embedding with Disentangled Graph Attention Network
• Graph Transformer for Recommendation
• Multi-Behavior Sequential Recommendation with Temporal Graph Transformer
• Improving Item Cold-start Recommendation via Model-agnostic Conditional Variational Autoencoder
• Invariant Collaborative Filtering to Popularity Distribution Shift
• DaisyRec 2.0: Benchmarking Recommendation for Rigorous Evaluation
• A Survey on Intent-aware Recommender Systems
• Context-aware Event Forecasting via Graph Disentanglement
• Intent Propagation Contrastive Collaborative Filtering
• Predicting Scientific Impact Through Diffusion, Conformity, and Contribution Disentanglement