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 ›› 2019, Vol. 45 ›› Issue (2): 195-201. doi: 10.19678/j.issn.1000-3428.0049732

• 人工智能及识别技术 • 上一篇    下一篇

基于相似度指标的社团划分算法

丁明珠,马英红,李云   

  1. 山东师范大学 管理科学与工程学院,济南 250014
  • 收稿日期:2017-12-19 出版日期:2019-02-15 发布日期:2019-02-15
  • 作者简介:丁明珠(1991—),女,硕士研究生,主研方向为大群体决策、复杂网络;马英红,教授、博士生导师;李云,硕士。
  • 基金资助:

    国家自然科学基金(71471106)。

Community Division Algorithm Based on Similarity Index

DING Mingzhu,MA Yinghong,LI Yun   

  1. School of Management Science and Engineering,Shandong Normal University,Jinan 250014,China
  • Received:2017-12-19 Online:2019-02-15 Published:2019-02-15

摘要:

为快速准确地找到复杂网络的社团划分结果,引入相似度指标,提出一种改进的社团划分算法。将网络节点拓扑结构信息作为节点的多维属性,在不改变网络连边的情况下,使输入网络转换为节点多属性网络,并定义节点之间的混合相似度与社团相似度,运用层次聚类思想得到最终的社团划分结果。在真实网络、计算机生成网络上的实验结果表明,该算法能够发现明显的社团结构,并且具有较高的社团划分准确率。

关键词: 复杂网络, 多属性, 社团划分, 模块度, 相似度

Abstract:

In order to find the result of complex network community division more accurately and quickly,an improved community division algorithm introduced similarity index is proposed.This paper extracts the network node topology information as a multi-dimensional attribute for each node,converts network into a multi attribute network by extracting topology information of nodes without changing the network side,defines the concepts of mixed node similarity and community similarity,and uses the hierarchical clustering idea to get the final community division result.This paper verifies that the algorithm can find the obvious community structure,improve the accuracy rate of community division by experiment on true network and computer generated network.

Key words: complex network, multiple attribute, community division, modularity, similarity

中图分类号: