作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2020, Vol. 46 ›› Issue (6): 94-102. doi: 10.19678/j.issn.1000-3428.0054210

• 人工智能与模式识别 • 上一篇    下一篇

融合节点属性与结构信息的子空间异常社区检测方法

赵琪琪1, 马慧芳1,2, 刘海姣1, 贾俊杰1   

  1. 1. 西北师范大学 计算机科学与工程学院, 兰州 730070;
    2. 桂林电子科技大学 广西可信软件重点实验室, 广西 桂林 541004
  • 收稿日期:2019-03-13 修回日期:2019-05-24 发布日期:2019-07-12
  • 作者简介:赵琪琪(1994-),女,硕士研究生,主研方向为异常社区检测、数据挖掘;马慧芳(通信作者),教授、博士;刘海姣,硕士研究生;贾俊杰,副教授、博士。
  • 基金资助:
    国家自然科学基金(61762078,61363058);广西可信软件重点实验室研究课题(kx202003);西北师范大学2019年青年教师科研能力提升计划重大项目(NWNU-LKQN2019-2);广西多源信息挖掘与安全重点实验室开放基金(MIMS18-08)。

Anomaly Community Detection Method via Subspace Combining Node Attribute and Structure Information

ZHAO Qiqi1, MA Huifang1,2, LIU Haijiao1, JIA Junjie1   

  1. 1. College of Computer Science and Engineering, Northwest Normal University, Lanzhou 730070, China;
    2. Guangxi Key Laboratory of Trusted Software, Guilin University of Electronic Technology, Guilin, Guangxi 541004, China
  • Received:2019-03-13 Revised:2019-05-24 Published:2019-07-12

摘要: 结合社区中的节点属性与结构信息,提出一种子空间异常社区检测方法。在待检测社区集合中,设计基于属性平均距离的子空间求解策略、基于负熵加权的子空间推断策略及子空间融合求解策略,挖掘每个社区的属性权重子空间,并根据社区结构关系定义社区质量评估模型量化社区质量分数,从而获得质量分数较低的异常社区集合。实验结果表明,该方法可以准确地发现异常社区,并且在人工网络和真实网络数据集上相比AMEN、SODA等检测方法具有更好的鲁棒性和可扩展性。

关键词: 社区集合, 属性信息, 结构信息, 异常社区检测, 子空间, 质量评估函数

Abstract: This paper proposes an anomaly community detection method via subspace by combining node attributes with structure information.First,in the given set of to-be-tested communities,the subspace solution strategy based on the average distance of attributes,the subspace inference strategy based on negative entropy weighting and the subspace fusion solution strategy are designed to excavate the attribute weight subspace of each community.Second,the quality assessment model is defined based on the community structure relationships to quantify the community quality scores.Finally,the set of anomaly communities with a low quality score is obtained.Experimental results show that the proposed method can accurately detect anomaly communities,and has better robustness and scalability than AMEN,SODA and other detection methods in artificial network and real network datasets.

Key words: community set, attribute information, structure information, anomaly community detection, subspace, quality assessment function

中图分类号: