Representation Learning for Attributed Multiplex Heterogeneous Network

Abstract

Network embedding (or graph embedding) has been widely used in many real-world applications. However, existing methods mainly focus on networks with single-typed nodes/edges and cannot scale well to handle large networks. Many real-world networks consist of billions of nodes and edges of multiple types, and each node is associated with different attributes. In this paper, we formalize the problem of embedding learning for the Attributed Multiplex Heterogeneous Network and propose a unified framework to address this problem. The framework supports both transductive and inductive learning. We also give the theoretical analysis of the proposed framework, showing its connection with previous works and proving its better expressiveness. We conduct systematical evaluations for the proposed framework on four different genres of challenging datasets: Amazon, YouTube, Twitter, and Alibaba. Experimental results demonstrate that with the learned embeddings from the proposed framework, we can achieve statistically significant improvements (e.g., 5.99-28.23% lift by F1 scores; p<<0.01, t-test) over previous state-of-the-art methods for link prediction. The framework has also been successfully deployed on the recommendation system of a worldwide leading e-commerce company, Alibaba Group. Results of the offline A/B tests on product recommendation further confirm the effectiveness and efficiency of the framework in practice.

Accepted to KDD 2019. Website: https://sites.google.com/view/gatne

References in corpus (15)

• Adam: A Method for Stochastic Optimization
• Distributed Representations of Words and Phrases and their Compositionality
• Community detection in graphs
• DeepWalk: Online Learning of Social Representations
• Semi-Supervised Classification with Graph Convolutional Networks
• LINE: Large-scale Information Network Embedding
• Inductive Representation Learning on Large Graphs
• A Structured Self-attentive Sentence Embedding
• Ups and Downs: Modeling the Visual Evolution of Fashion Trends with One-Class Collaborative Filtering
• Variational Graph Auto-Encoders
• PTE: Predictive Text Embedding through Large-scale Heterogeneous Text Networks
• Network Embedding as Matrix Factorization: Unifying DeepWalk, LINE, PTE, and node2vec
• Attributed Social Network Embedding
• NetSMF: Large-Scale Network Embedding as Sparse Matrix Factorization
• mvn2vec: Preservation and Collaboration in Multi-View Network Embedding

Cited by in corpus (61)

• MAGNN: Metapath Aggregated Graph Neural Network for Heterogeneous Graph Embedding
• Fast Parallel Hypertree Decompositions in Logarithmic Recursion Depth
• Graph Meta Network for Multi-Behavior Recommendation
• HDMI: High-order Deep Multiplex Infomax
• Multiplex Behavioral Relation Learning for Recommendation via Memory Augmented Transformer Network
• Attention-based graph neural networks: a survey
• Multiplex Heterogeneous Graph Convolutional Network
• Heterogeneous Graph Representation Learning with Relation Awareness
• Multi-behavior Self-supervised Learning for Recommendation
• GraphSAIL: Graph Structure Aware Incremental Learning for Recommender Systems
• Multi-Behavior Graph Neural Networks for Recommender System
• TIMME: Twitter Ideology-detection via Multi-task Multi-relational Embedding
• mvn2vec: Preservation and Collaboration in Multi-View Network Embedding
• Graph Communal Contrastive Learning
• Multilayer Network Analysis for Improved Credit Risk Prediction
• Are we really making much progress? Revisiting, benchmarking, and refining heterogeneous graph neural networks
• xFraud: Explainable Fraud Transaction Detection
• Recommender systems based on graph embedding techniques: A comprehensive review
• Graph Contrastive Learning with Generative Adversarial Network
• Medical Entity Disambiguation Using Graph Neural Networks
• Heterogeneous Network Representation Learning: A Unified Framework with Survey and Benchmark
• X-GOAL: Multiplex Heterogeneous Graph Prototypical Contrastive Learning
• PANE: scalable and effective attributed network embedding
• Tripartite Heterogeneous Graph Propagation for Large-scale Social Recommendation
• Convolutional Learning on Multigraphs
• Multi-View Collaborative Network Embedding
• Incorporating Attributes and Multi-Scale Structures for Heterogeneous Graph Contrastive Learning
• Controllable Multi-Interest Framework for Recommendation
• TabGNN: Multiplex Graph Neural Network for Tabular Data Prediction
• Hierarchical Aggregations for High-Dimensional Multiplex Graph Embedding
• Recommending on graphs: a comprehensive review from a data perspective
• CoarSAS2hvec: Heterogeneous Information Network Embedding with Balanced Network Sampling
• RGAT: A Deeper Look into Syntactic Dependency Information for Coreference Resolution
• Modeling Dynamic Heterogeneous Network for Link Prediction using Hierarchical Attention with Temporal RNN
• RetaGNN: Relational Temporal Attentive Graph Neural Networks for Holistic Sequential Recommendation
• Zoo Guide to Network Embedding
• DeepTrax: Embedding Graphs of Financial Transactions
• Multi-Relational Graph based Heterogeneous Multi-Task Learning in Community Question Answering
• Unsupervised Multiplex Graph Learning with Complementary and Consistent Information
• Knowledge-aware Coupled Graph Neural Network for Social Recommendation
• CogDL: A Comprehensive Library for Graph Deep Learning
• Self-Supervised Learning of Contextual Embeddings for Link Prediction in Heterogeneous Networks
• Lifelong Property Price Prediction: A Case Study for the Toronto Real Estate Market
• Leveraging Meta-path Contexts for Classification in Heterogeneous Information Networks
• Concept-Aware Denoising Graph Neural Network for Micro-Video Recommendation
• Graph4Rec: A Universal Toolkit with Graph Neural Networks for Recommender Systems
• Deep Unified Representation for Heterogeneous Recommendation
• Deep Contrastive Multiview Network Embedding
• Uniting Heterogeneity, Inductiveness, and Efficiency for Graph Representation Learning
• Path-Based Reasoning over Heterogeneous Networks for Recommendation via Bidirectional Modeling
• Document Structure aware Relational Graph Convolutional Networks for Ontology Population
• A Geometric Perspective for High-Dimensional Multiplex Graphs
• Dual prototype attentive graph network for cross-market recommendation
• Automated Feature-Topic Pairing: Aligning Semantic and Embedding Spaces in Spatial Representation Learning
• Grouping Search Results with Product Graphs in E-commerce Platforms
• Heterogeneous Graph Neural Network with Multi-view Representation Learning
• Network representation learning: A macro and micro view
• A Multi-Semantic Metapath Model for Large Scale Heterogeneous Network Representation Learning
• Hyperbolic Multiplex Network Embedding with Maps of Random Walk
• Graph Representation Learning Network via Adaptive Sampling
• Multiplex Bipartite Network Embedding using Dual Hypergraph Convolutional Networks