Journal of Computer Applications ›› 2019, Vol. 39 ›› Issue (12): 3415-3419.DOI: 10.11772/j.issn.1001-9081.2019071281

• The 17th China Conference on Machine Learning (CCML 2019) •     Next Articles

Graph convolutional network model using neighborhood selection strategy

CHEN Kejia1,2, YANG Zeyu1, LIU Zheng1, LU Hao1,2   

  1. 1. School of Computer Science, Nanjing University of Posts and Telecommunications, Nanjing Jiangsu 210046, China;
    2. Jiangsu Key Laboratory of Big Data Security and Intelligent Processing, Nanjing Jiangsu 210046, China
  • Received:2019-04-29 Revised:2019-08-07 Online:2019-12-10 Published:2019-08-26
  • Contact: 陈可佳
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61603197, 61772284).

基于邻域选择策略的图卷积网络模型

陈可佳1,2, 杨泽宇1, 刘峥1, 鲁浩1,2   

  1. 1. 南京邮电大学 计算机学院, 南京 210046;
    2. 江苏省大数据挖掘与智能计算重点实验室, 南京 210046
  • 作者简介:陈可佳(1980-),女,江苏淮安人,副教授,博士,CCF会员,主要研究方向:数据挖掘、复杂网络分析;杨泽宇(1994-),男,山西晋中人,硕士研究生,主要研究方向:图卷积网络;刘峥(1980-),男,江苏南京人,讲师,博士,CCF会员,主要研究方向:图数据挖掘;鲁浩(1995-),男,河南濮阳人,硕士研究生,主要研究方向:图卷积网络、网络表示学习。
  • 基金资助:
    国家自然科学基金资助项目(61603197,61772284)。

Abstract: The composition of neighborhoods is crucial for the spatial domain-based Graph Convolutional Network (GCN) model. To solve the problem that the structural influence is not considered in the neighborhood ordering of nodes in the model, a novel neighborhood selection strategy was proposed to obtain an improved GCN model. Firstly, the structurally important neighborhoods were collected for each node and the core neighborhoods were selected hierarchically. Secondly, the features of the nodes and their core neighborhoods were organized into a matrix. Finally, the matrix was sent to deep Convolutional Neural Network (CNN) for semi-supervised learning. The experimental results on Cora, Citeseer and Pubmed citation network datasets show that, the proposed model has a better accuracy in node classification tasks than the model based on classical graph embedding and four state-of-the-art GCN models. As a spatial domain-based GCN, the proposed model can be effectively applied to the learning tasks of large-scale networks.

Key words: Graph Convolutional Network (GCN), neighborhood selection strategy, graph embedding, node classification, semi-supervised learning

摘要: 邻域的组成对于基于空间域的图卷积网络(GCN)模型有至关重要的作用。针对模型中节点邻域排序未考虑结构影响力的问题,提出了一种新的邻域选择策略,从而得到改进的GCN模型。首先,为每个节点收集结构重要的邻域并进行层级选择得到核心邻域;然后,将节点及其核心邻域的特征组成有序的矩阵形式;最后,送入深度卷积神经网络(CNN)进行半监督学习。节点分类任务的实验结果表明,该模型在Cora、Citeseer和Pubmed引文网络数据集中的节点分类准确性均优于基于经典图嵌入的节点分类模型以及四种先进的GCN模型。作为一种基于空间域的GCN,该模型能有效运用于大规模网络的学习任务。

关键词: 图卷积网络, 邻域选择策略, 图嵌入, 节点分类, 半监督学习

CLC Number: