Journal of Frontiers of Computer Science and Technology ›› 2022, Vol. 16 ›› Issue (12): 2788-2796.DOI: 10.3778/j.issn.1673-9418.2104116

• Artificial Intelligence • Previous Articles     Next Articles

Method of K-order Graph Convolution Attribute Network Community Detection

CHEN Jie1, ZHANG Erming1, WANG Qianqian2, ZHAO Shu1,+(), ZHANG Yanping1   

  1. 1. College of Computer Science and Technology, Anhui University, Hefei 230601, China
    2. College of International Education, Anhui University, Hefei 230601, China
  • Received:2021-05-08 Revised:2021-06-25 Online:2022-12-01 Published:2021-06-08
  • About author:CHEN Jie, born in 1982, Ph.D., lecturer, mem-ber of CCF. Her research interests include in-telligent computing, machine learning, etc.
    ZHANG Erming, born in 1995, M.S. candi-date. His research interests include intelligent computing, machine learning, etc.
    WANG Qianqian, born in 1982, M.S., lecturer, member of CAAI. Her research interests include granular computing and machine learning.
    ZHAO Shu, born in 1979, Ph.D., professor, Ph.D. supervisor, senior member of CCF. Her research interests include granular computing, quotient space theory and machine learning.
    ZHANG Yanping, born in 1962, professor, Ph.D. supervisor. Her research interests include in-telligent computing, quotient space, machine lear-ning, intelligent information processing, etc.
  • Supported by:
    National Key Research and Development Program Subtopic of China(2017YFB1401903);National Natural Science Foundation of China(61876001)

K阶图卷积属性网络社团检测方法

陈洁1, 张二明1, 王倩倩2, 赵姝1,+(), 张燕平1   

  1. 1.安徽大学 计算机科学与技术学院,合肥 230601
    2.安徽大学 国际教育学院,合肥 230601
  • 通讯作者: +E-mail: zhaoshuzs2002@hotmail.com
  • 作者简介:陈洁(1982—),女,安徽巢湖人,博士,讲师,CCF会员,主要研究方向为智能计算、机器学习等。
    张二明(1995—),男,安徽淮北人,硕士研究生,主要研究方向为智能计算、机器学习等。
    王倩倩(1982—),女,安徽六安人,硕士,讲师,CAAI会员,主要研究方向为粒计算、机器学习。
    赵姝(1979—),女,安徽巢湖人,博士,教授,博士生导师,CCF高级会员,主要研究方向为粒计算、商空间理论、机器学习。
    张燕平(1962—),女,安徽合肥人,教授,博士生导师,主要研究方向为智能计算、商空间、机器学习、智能信息处理等。
  • 基金资助:
    国家重点研发计划子课题(2017YFB1401903);国家自然科学基金(61876001)

Abstract:

Mining community structure in attribute network is helpful to further analysis of network nodes, which has important practical significance. Graph convolution neural network can effectively embed the structural information of attribute network and obtain the feature representation of nodes. Based on this, the community structure with good performance can be obtained. However, most of the existing graph convolution methods use fixed low-order graph convolution, only consider the first-order or second-order neighbors of each node, don’t make full use of the node relationship and ignore the diversity of network structure. In addition, the sparsity of the original network structure can’t be overcome, which will reduce the performance of community detection. In order to solve the above problems, this paper proposes a K-order graph convolution community detection method (KGCN) which combines attribute information and structure information. This method can effectively overcome the sparsity of the original network and use the high-order structure of nodes for community detection. Firstly, the original network is reconstructed according to the attribute information of nodes to alleviate the sparsity of the original network. Secondly, considering the high-order structure correlation, the K-order graph convolution encoder is used to encode the nodes to obtain the feature representation of the nodes. Finally, community detection is carried out using spectral clustering algorithm. Experimental results show that the KGCN method achieves better community detection results than the existing algorithms on four real datasets.

Key words: community detection, K-order graph convolution, attribute information, sparsity

摘要:

挖掘属性网络中的社团结构有助于对网络节点进一步分析,具有重要的现实意义。图卷积神经网络能够有效地将属性网络的结构信息进行嵌入,获取节点的特征表示,从而可获得性能良好的社团结构。然而,现有图卷积方法大多使用固定的低阶图卷积,只考虑每个节点一阶或二阶内的邻居,没有充分利用节点关系,忽略了网络结构的多样性。另外原始网络结构的稀疏性无法克服,会降低社团检测的性能。为解决上述问题,提出一种融合属性信息与结构信息的K阶图卷积社团检测方法(KGCN),该方法可以有效地克服原始网络的稀疏性并利用节点的高阶结构进行社团检测。首先根据节点的属性信息对原始网络进行重构,缓解原始网络结构的稀疏性;其次考虑到高阶结构关联,采用K阶图卷积编码器对节点进行编码,获得节点的特征表示;最后使用谱聚类算法进行社团检测。实验结果表明,在四个真实数据集上,相比现有算法,KGCN方法取得更好的社团检测结果。

关键词: 社团检测, K阶图卷积, 属性信息, 稀疏性

CLC Number: