Multiplex Heterogeneous Graph Convolutional Network

Abstract

Heterogeneous graph convolutional networks have gained great popularity in tackling various network analytical tasks on heterogeneous network data, ranging from link prediction to node classification. However, most existing works ignore the relation heterogeneity with multiplex network between multi-typed nodes and different importance of relations in meta-paths for node embedding, which can hardly capture the heterogeneous structure signals across different relations. To tackle this challenge, this work proposes a Multiplex Heterogeneous Graph Convolutional Network (MHGCN) for heterogeneous network embedding. Our MHGCN can automatically learn the useful heterogeneous metapath interactions of different lengths in multiplex heterogeneous networks through multi-layer convolution aggregation. Additionally, we effectively integrate both multi-relation structural signals and attribute semantics into the learned node embeddings with both unsupervised and semi-supervised learning paradigms. Extensive experiments on five real-world datasets with various network analytical tasks demonstrate the significant superiority of MHGCN against state-of-the-art embedding baselines in terms of all evaluation metrics. The source code of our method is available at: https://github.com/NSSSJSS/MHGCN. CCS CONCEPTS• Mathematics of computing → Graph algorithms; • Computing methodologies → Learning latent representations.

Random Walk on Multiple Networks

“…Recently, multiple networks have drawn increasing attention in the literature due to their capability in describing graph-structure data from different domains [26], [40], [43], [14], [61], [59], [60], [56]. Under the multiple network setting, a wide range of graph mining tasks have been extended to support more realistic real-life applications, including node representation learning [40], [26], [14], [60], node clustering [11], [29], [43], [33], and link prediction [61], [59]. For example, Multiplex Graph Neural Network is proposed to tackle the multi-behavior recommendation problem [61].…”

“…For example, Multiplex Graph Neural Network is proposed to tackle the multi-behavior recommendation problem [61]. In [59], the authors consider the relational heterogeneity within multiplex networks and propose a multiplex heterogeneous graph convolutional network (MHGCN) to learn node representations in hetero- One-hot vector with only one value-1 entry for uq x (t) i Node visit. prob.…”

Multi-Behavior Graph Neural Networks for Recommender System

“…These neural network models generally focus on single type of user interaction behaviors over items, during the vectorized representation procedure of users and items. However, in reallife applications, items are often interacted by users with diverse ways [7], [54], [51]. For example, users can view, tagas-favourite and purchase different products in E-commerce platforms.…”

Graph Structure Learning on User Mobility Data for Social Relationship Inference

“…They work poorly for inactive users since they fail to solve the data sparsity challenge of inactive users. Recently, graph embedding (Perozzi, Al-Rfou, and Skiena 2014;Grover and Leskovec 2016;Liu et al 2020) and graph convolutional networks (GCNs) (Kipf and Welling 2017; Veličković et al 2018;Hamilton, Ying, and Leskovec 2017;Liu et al 2021;Yu et al 2022) have been developed for learning node representations in graph-structured data. A line of works based on network embedding and GCNs have been proposed, which no longer require hand-crafted features, but treat users as nodes and model the meeting events between users as a homogeneous graph (Yu, Wang, and Li 2018) or a heterogeneous graph (Backes et al 2017;Wu et al 2019).…”