KGAT: Knowledge Graph Attention Network for Recommendation

Abstract

To provide more accurate, diverse, and explainable recommendation, it is compulsory to go beyond modeling user-item interactions and take side information into account. Traditional methods like factorization machine (FM) cast it as a supervised learning problem, which assumes each interaction as an independent instance with side information encoded. Due to the overlook of the relations among instances or items (e.g., the director of a movie is also an actor of another movie), these methods are insufficient to distill the collaborative signal from the collective behaviors of users. In this work, we investigate the utility of knowledge graph (KG), which breaks down the independent interaction assumption by linking items with their attributes. We argue that in such a hybrid structure of KG and user-item graph, high-order relations --- which connect two items with one or multiple linked attributes --- are an essential factor for successful recommendation. We propose a new method named Knowledge Graph Attention Network (KGAT) which explicitly models the high-order connectivities in KG in an end-to-end fashion. It recursively propagates the embeddings from a node's neighbors (which can be users, items, or attributes) to refine the node's embedding, and employs an attention mechanism to discriminate the importance of the neighbors. Our KGAT is conceptually advantageous to existing KG-based recommendation methods, which either exploit high-order relations by extracting paths or implicitly modeling them with regularization. Empirical results on three public benchmarks show that KGAT significantly outperforms state-of-the-art methods like Neural FM and RippleNet. Further studies verify the efficacy of embedding propagation for high-order relation modeling and the interpretability benefits brought by the attention mechanism.

KDD 2019 research track

References in corpus (19)

• Adam: A Method for Stochastic Optimization
• Semi-Supervised Classification with Graph Convolutional Networks
• Inductive Representation Learning on Large Graphs
• BPR: Bayesian Personalized Ranking from Implicit Feedback
• Neural Graph Collaborative Filtering
• Ups and Downs: Modeling the Visual Evolution of Fashion Trends with One-Class Collaborative Filtering
• RippleNet: Propagating User Preferences on the Knowledge Graph for Recommender Systems
• Graph Convolutional Matrix Completion
• xDeepFM: Combining Explicit and Implicit Feature Interactions for Recommender Systems
• Representation Learning on Graphs with Jumping Knowledge Networks
• Unifying Knowledge Graph Learning and Recommendation: Towards a Better Understanding of User Preferences
• NAIS: Neural Attentive Item Similarity Model for Recommendation
• DeepInf: Social Influence Prediction with Deep Learning
• Learning over Knowledge-Base Embeddings for Recommendation
• Learning Heterogeneous Knowledge Base Embeddings for Explainable Recommendation
• Wide & Deep Learning for Recommender Systems
• Item Silk Road: Recommending Items from Information Domains to Social Users
• Neural Collective Entity Linking
• KB4Rec: A Dataset for Linking Knowledge Bases with Recommender Systems

Cited by in corpus (205)

• Neural Graph Collaborative Filtering
• A Survey on Knowledge Graphs: Representation, Acquisition and Applications
• Self-supervised Graph Learning for Recommendation
• Disentangled Graph Collaborative Filtering
• Learning Intents behind Interactions with Knowledge Graph for Recommendation
• Knowledge Graph Contrastive Learning for Recommendation
• A Survey on Accuracy-oriented Neural Recommendation: From Collaborative Filtering to Information-rich Recommendation
• LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation
• When Large Language Models Meet Personalization: Perspectives of Challenges and Opportunities
• Deep Entity Matching with Pre-Trained Language Models
• Heterogeneous Graph Contrastive Learning for Recommendation
• Multi-level Cross-view Contrastive Learning for Knowledge-aware Recommender System
• Multi-Modal Self-Supervised Learning for Recommendation
• Graph Meta Network for Multi-Behavior Recommendation
• Contrastive Meta Learning with Behavior Multiplicity for Recommendation
• SumGNN: Multi-typed Drug Interaction Prediction via Efficient Knowledge Graph Summarization
• Research Commentary on Recommendations with Side Information: A Survey and Research Directions
• Knowledge Graph Self-Supervised Rationalization for Recommendation
• DisenHAN: Disentangled Heterogeneous Graph Attention Network for Recommendation
• Reinforced Negative Sampling over Knowledge Graph for Recommendation
• Graph Neural Networks: Taxonomy, Advances and Trends
• Disentangled Contrastive Collaborative Filtering
• Knowledge Graphs in Manufacturing and Production: A Systematic Literature Review
• Multiplex Behavioral Relation Learning for Recommendation via Memory Augmented Transformer Network
• Graph Enhanced Representation Learning for News Recommendation
• DGCN: Diversified Recommendation with Graph Convolutional Networks
• BERT Based Clinical Knowledge Extraction for Biomedical Knowledge Graph Construction and Analysis
• Learning on Attribute-Missing Graphs
• Automated Self-Supervised Learning for Recommendation
• STP-UDGAT: Spatial-Temporal-Preference User Dimensional Graph Attention Network for Next POI Recommendation
• Attention-based graph neural networks: a survey
• Social Recommendation with Self-Supervised Metagraph Informax Network
• KG4Vis: A Knowledge Graph-Based Approach for Visualization Recommendation
• MetaKG: Meta-learning on Knowledge Graph for Cold-start Recommendation
• Multi-Behavior Recommendation with Cascading Graph Convolution Networks
• Knowledge Enhancement for Contrastive Multi-Behavior Recommendation
• IMF: Interactive Multimodal Fusion Model for Link Prediction
• KRED: Knowledge-Aware Document Representation for News Recommendations
• Improving Knowledge-aware Recommendation with Multi-level Interactive Contrastive Learning
• CAFE: Coarse-to-Fine Neural Symbolic Reasoning for Explainable Recommendation
• Solving Cold Start Problem in Recommendation with Attribute Graph Neural Networks
• A Survey on The Expressive Power of Graph Neural Networks
• Graph Convolution Machine for Context-aware Recommender System
• Construction of Knowledge Graphs: State and Challenges
• Multi-behavior Self-supervised Learning for Recommendation
• Time-aware Path Reasoning on Knowledge Graph for Recommendation
• How to Retrain Recommender System? A Sequential Meta-Learning Method
• Representation Learning for Dynamic Graphs: A Survey
• Learning Robust Recommenders through Cross-Model Agreement
• ATBRG: Adaptive Target-Behavior Relational Graph Network for Effective Recommendation
• Large-scale Personalized Video Game Recommendation via Social-aware Contextualized Graph Neural Network
• Joint Multi-grained Popularity-aware Graph Convolution Collaborative Filtering for Recommendation
• HS-GCN: Hamming Spatial Graph Convolutional Networks for Recommendation
• Are we really making much progress? Revisiting, benchmarking, and refining heterogeneous graph neural networks
• Invariant Collaborative Filtering to Popularity Distribution Shift
• FairSR: Fairness-aware Sequential Recommendation through Multi-Task Learning with Preference Graph Embeddings
• Semi-Supervised Variational Reasoning for Medical Dialogue Generation
• Post Processing Recommender Systems with Knowledge Graphs for Recency, Popularity, and Diversity of Explanations
• Self-Supervised Hyperboloid Representations from Logical Queries over Knowledge Graphs
• Causality-based CTR Prediction using Graph Neural Networks
• Neural Graph Matching based Collaborative Filtering
• GDSRec: Graph-Based Decentralized Collaborative Filtering for Social Recommendation
• Attention Is Not the Only Choice: Counterfactual Reasoning for Path-Based Explainable Recommendation
• Recommender systems based on graph embedding techniques: A comprehensive review
• VRKG4Rec: Virtual Relational Knowledge Graphs for Recommendation
• Let Me Do It For You: Towards LLM Empowered Recommendation via Tool Learning
• RecBole: Towards a Unified, Comprehensive and Efficient Framework for Recommendation Algorithms
• The Expressive Power of Graph Neural Networks: A Survey
• Knowledge-refined Denoising Network for Robust Recommendation
• Deep Learning on Knowledge Graph for Recommender System: A Survey
• paper2repo: GitHub Repository Recommendation for Academic Papers
• Multi-domain Recommendation with Embedding Disentangling and Domain Alignment
• Graph Learning Approaches to Recommender Systems: A Review
• $\alpha$-Satellite: An AI-driven System and Benchmark Datasets for Hierarchical Community-level Risk Assessment to Help Combat COVID-19
• DaisyRec 2.0: Benchmarking Recommendation for Rigorous Evaluation
• A Survey on Knowledge Graph-Based Recommender Systems
• Domain Disentanglement with Interpolative Data Augmentation for Dual-Target Cross-Domain Recommendation
• Knowledge Enhanced Multi-intent Transformer Network for Recommendation
• Uncovering CWE-CVE-CPE Relations with Threat Knowledge Graphs
• Multivariate and Propagation Graph Attention Network for Spatial-Temporal Prediction with Outdoor Cellular Traffic
• Schema-aware Reference as Prompt Improves Data-Efficient Knowledge Graph Construction
• Rank List Sensitivity of Recommender Systems to Interaction Perturbations
• INMO: A Model-Agnostic and Scalable Module for Inductive Collaborative Filtering
• Sequential Recommendation with Graph Neural Networks
• HeteGCN: Heterogeneous Graph Convolutional Networks for Text Classification
• Triangle Graph Interest Network for Click-through Rate Prediction
• Rethinking Collaborative Metric Learning: Toward an Efficient Alternative without Negative Sampling
• User Consented Federated Recommender System Against Personalized Attribute Inference Attack
• GIKT: A Graph-based Interaction Model for Knowledge Tracing
• Fairness-Aware Explainable Recommendation over Knowledge Graphs
• On Provable Benefits of Depth in Training Graph Convolutional Networks
• M2GNN: Metapath and Multi-interest Aggregated Graph Neural Network for Tag-based Cross-domain Recommendation
• Challenging the Myth of Graph Collaborative Filtering: a Reasoned and Reproducibility-driven Analysis
• Comprehending Knowledge Graphs with Large Language Models for Recommender Systems
• HIEN: Hierarchical Intention Embedding Network for Click-Through Rate Prediction
• Collaborative Cross-modal Fusion with Large Language Model for Recommendation
• DA-GCN: A Domain-aware Attentive Graph Convolution Network for Shared-account Cross-domain Sequential Recommendation
• Urban Computing in the Era of Large Language Models
• Recommending on graphs: a comprehensive review from a data perspective
• Hierarchical Fashion Graph Network for Personalized Outfit Recommendation
• Graph Neural News Recommendation with Long-term and Short-term Interest Modeling
• A Light Heterogeneous Graph Collaborative Filtering Model using Textual Information
• Unify Local and Global Information for Top-$N$ Recommendation
• PARSRec: Explainable Personalized Attention-fused Recurrent Sequential Recommendation Using Session Partial Actions
• NudgeRank: Digital Algorithmic Nudging for Personalized Health
• Cross-Network Learning with Partially Aligned Graph Convolutional Networks
• Graph Learning based Recommender Systems: A Review
• Dual Policy Learning for Aggregation Optimization in Graph Neural Network-based Recommender Systems
• Inhomogeneous Social Recommendation with Hypergraph Convolutional Networks
• FedRKG: A Privacy-preserving Federated Recommendation Framework via Knowledge Graph Enhancement
• COOKIE: A Dataset for Conversational Recommendation over Knowledge Graphs in E-commerce
• RecKG: Knowledge Graph for Recommender Systems
• Try This Instead: Personalized and Interpretable Substitute Recommendation
• LightKG: Efficient Knowledge-Aware Recommendations with Simplified GNN Architecture
• RetaGNN: Relational Temporal Attentive Graph Neural Networks for Holistic Sequential Recommendation
• DebiasedRec: Bias-aware User Modeling and Click Prediction for Personalized News Recommendation
• Graph Neural Pre-training for Enhancing Recommendations using Side Information
• Constant Time Graph Neural Networks
• Implicit Session Contexts for Next-Item Recommendations
• LogicRec: Recommendation with Users' Logical Requirements
• LightSAGE: Graph Neural Networks for Large Scale Item Retrieval in Shopee's Advertisement Recommendation
• A Knowledge Graph based Approach for Mobile Application Recommendation
• Multi-Graph based Multi-Scenario Recommendation in Large-scale Online Video Services
• TinyKG: Memory-Efficient Training Framework for Knowledge Graph Neural Recommender Systems
• Continuous-Time Sequential Recommendation with Temporal Graph Collaborative Transformer
• HGCH: A Hyperbolic Graph Convolution Network Model for Heterogeneous Collaborative Graph Recommendation
• Heterogeneity-aware Cross-school Electives Recommendation: a Hybrid Federated Approach
• Bayesian Knowledge-driven Critiquing with Indirect Evidence
• Knowledge-aware Coupled Graph Neural Network for Social Recommendation
• AutoETER: Automated Entity Type Representation for Knowledge Graph Embedding
• Document Modeling with Graph Attention Networks for Multi-grained Machine Reading Comprehension
• Dual Preference Distribution Learning for Item Recommendation
• Knowledge-Enhanced Recommendation with User-Centric Subgraph Network
• KG4RecEval: Does Knowledge Graph Really Matter for Recommender Systems?
• Pre-Training Graph Neural Networks for Cold-Start Users and Items Representation
• Metapath- and Entity-aware Graph Neural Network for Recommendation
• Improving Location Recommendation with Urban Knowledge Graph
• Meta-GPS++: Enhancing Graph Meta-Learning with Contrastive Learning and Self-Training
• LaSER: Language-Specific Event Recommendation
• Start from Zero: Triple Set Prediction for Automatic Knowledge Graph Completion
• DyFormer: A Scalable Dynamic Graph Transformer with Provable Benefits on Generalization Ability
• Heterogeneous Graph Collaborative Filtering
• Social Explorative Attention based Recommendation for Content Distribution Platforms
• Contextualized Graph Attention Network for Recommendation with Item Knowledge Graph
• Exploring Data Splitting Strategies for the Evaluation of Recommendation Models
• Syndrome-aware Herb Recommendation with Multi-Graph Convolution Network
• Type-augmented Relation Prediction in Knowledge Graphs
• A^2-GCN: An Attribute-aware Attentive GCN Model for Recommendation
• Faithfully Explainable Recommendation via Neural Logic Reasoning
• Graph Attention Collaborative Similarity Embedding for Recommender System
• Interactive Recommender System via Knowledge Graph-enhanced Reinforcement Learning
• Minimal Variance Sampling with Provable Guarantees for Fast Training of Graph Neural Networks
• Graph-Based Recommendation System Enhanced with Community Detection
• Assisted Knowledge Graph Authoring: Human-Supervised Knowledge Graph Construction from Natural Language
• Explainable Recommender Systems via Resolving Learning Representations
• Multi-representations Space Separation based Graph-level Anomaly-aware Detection
• Group-Buying Recommendation for Social E-Commerce
• Alleviating Cold-Start Problems in Recommendation through Pseudo-Labelling over Knowledge Graph
• InsertGNN: Can Graph Neural Networks Outperform Humans in TOEFL Sentence Insertion Problem?
• Interest-aware Message-Passing GCN for Recommendation
• Uniting Heterogeneity, Inductiveness, and Efficiency for Graph Representation Learning
• Reproducibility and Artifact Consistency of the SIGIR 2022 Recommender Systems Papers Based on Message Passing
• Topic-Aware Knowledge Graph with Large Language Models for Interoperability in Recommender Systems
• MixDec Sampling: A Soft Link-based Sampling Method of Graph Neural Network for Recommendation
• Concept-Aware Denoising Graph Neural Network for Micro-Video Recommendation
• The Graph-Based Behavior-Aware Recommendation for Interactive News
• Single-Layer Graph Convolutional Networks For Recommendation
• The Limits of Graph Samplers for Training Inductive Recommender Systems: Extended results
• SceneRec: Scene-Based Graph Neural Networks for Recommender Systems
• Rule-Guided Graph Neural Networks for Recommender Systems
• SPARK: Adaptive Low-Rank Knowledge Graph Modeling in Hybrid Geometric Spaces for Recommendation
• Multi-Level Attention Pooling for Graph Neural Networks: Unifying Graph Representations with Multiple Localities
• DRGraph: An Efficient Graph Layout Algorithm for Large-scale Graphs by Dimensionality Reduction
• A Unified Framework for Cross-Domain and Cross-System Recommendations
• Augmenting the User-Item Graph with Textual Similarity Models
• Personalized Hashtag Recommendation for Micro-videos
• printf: Preference Modeling Based on User Reviews with Item Images and Textual Information via Graph Learning
• Graph-based Recommendation for Sparse and Heterogeneous User Interactions
• Can Offline Metrics Measure Explanation Goals? A Comparative Survey Analysis of Offline Explanation Metrics in Recommender Systems
• Rumor Detection on Twitter with Claim-Guided Hierarchical Graph Attention Networks
• On the Importance of Sampling in Training GCNs: Tighter Analysis and Variance Reduction
• Hierarchical User Intent Graph Network forMultimedia Recommendation
• RUEL: Retrieval-Augmented User Representation with Edge Browser Logs for Sequential Recommendation
• DCDIR: A Deep Cross-Domain Recommendation System for Cold Start Users in Insurance Domain
• Curvature Graph Neural Network
• Efficient Data-specific Model Search for Collaborative Filtering
• UGRec: Modeling Directed and Undirected Relations for Recommendation
• Rating and aspect-based opinion graph embeddings for explainable recommendations
• Dual Graph enhanced Embedding Neural Network for CTR Prediction
• A Concept Knowledge-Driven Keywords Retrieval Framework for Sponsored Search
• Leveraging Tripartite Interaction Information from Live Stream E-Commerce for Improving Product Recommendation
• URIR: Recommendation algorithm of user RNN encoder and item encoder based on knowledge graph
• KATRec: Knowledge Aware aTtentive Sequential Recommendations
• Graphing else matters: exploiting aspect opinions and ratings in explainable graph-based recommendations
• SAHDL: Sparse Attention Hypergraph Regularized Dictionary Learning
• Mining Implicit Entity Preference from User-Item Interaction Data for Knowledge Graph Completion via Adversarial Learning
• Dual prototype attentive graph network for cross-market recommendation
• M-$LL{M}^3$REC: A Motivation-Aware User-Item Interaction Framework for Enhancing Recommendation Accuracy with LLMs
• Node Attribute Completion in Knowledge Graphs with Multi-Relational Propagation
• GRCN: Graph-Refined Convolutional Network for Multimedia Recommendation with Implicit Feedback
• Skewness Ranking Optimization for Personalized Recommendation
• Conditional Attention Networks for Distilling Knowledge Graphs in Recommendation
• Quaternion-Based Graph Convolution Network for Recommendation
• Pre-training Recommender Systems via Reinforced Attentive Multi-relational Graph Neural Network
• Dual Graph Embedding for Object-Tag LinkPrediction on the Knowledge Graph