Abstract

Node classification, as a central task in the graph data analysis, has been studied extensively with network embedding technique for single-layer graph network. However, there are some obstacles when extending the single-layer network embedding technique to the attributed multiplex network. The classification of a given node in the attributed multiplex network must consider the network structure in different dimensions, as well as rich node attributes, and correlations among the different dimensions. Moreover, the distance node context information of a given node in each dimension will also affect the classification of the given node. In this study, a novel network embedding approach for the node classification of attributed multiplex networks using random walk and graph convolutional networks (AMRG) is proposed. A random walk network embedding technique was used to extract distant node information and the results are considered as pre-trained node features to be concatenated with the original node features inputted into the graph convolutional networks (GCNs) to learn node representations for each dimension. Besides, the consensus regularization is introduced to capture the similarities among different dimensions, and the learnable neural network parameters of GCNs for different dimensions are also constrained by the regularization mechanism to improve the correlations. As well as an attention mechanism is explored to infer the importance for a given node in different dimensions. Extensive experiments demonstrated that our proposed technique outperforms many competitive baselines on several real-world multiplex network datasets.

Highlights

  • Node classification [1,2] is a basic and central task in the graph data analysis, such as the user division in social networks [3], the paper classification in citation network [4]

  • We provide a novel node classification methodology for attributed multiplex networks using random walk network embedding and graph convolutional networks (AMRG), which can fuse the node attributes and capture distant node context information

  • Our proposed technique is primarily composed of four components: 1) the pre-training of node features using a random walk network embedding method, 2) dimension specific node embedding using a graph convolutional networks (GCNs) model, 3) cross dimension modeling terms, and 4) an attention-based mechanism used to generate global node representations by integrating embeddings of different dimensions

Read more

Summary

INTRODUCTION

Node classification [1,2] is a basic and central task in the graph data analysis, such as the user division in social networks [3], the paper classification in citation network [4]. Previous techniques such as PMNE [19], MELL [20], MVE [21] and MNE [11] have learned to integrate node embedding information from different dimensions in multiplex network representations. The primary challenge for node classification of multiplex network is designing a model to extract the node information, oriented by the downstream node classification task, capable of generating a comprehensive embedding (consensus) that considers node attributes, their interaction and similarities among different dimensions, the corresponding degree of importance in diverse dimension networks, and the distance node context information. Used to adaptively learn the importance weights of a given node in different dimensional, prior to integrating the node embedding results from different dimensions to generate global consensus node representations This model can be trained end-to-end oriented by the downstream node classification task.

Single-Layer Network Embedding
Multiplex Network Embedding
Problem Statement and Framework
Capturing Distance Neighboring Node Information
Dimension Specific Node Embedding With GCN
Cross Dimension Modeling
Attention Mechanisms for Fusing Different Dimensions
Optimization Objective
Time Complexity
Experimental Setup
Node Classification
Analysis of Attention Mechanisms
Analysis of Variants
Analysis of Key Factor
Visualization
CONCLUSION
Findings
DATA AVAILABILITY STATEMENT

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.