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 ›› 2013, Vol. 39 ›› Issue (7): 242-246. doi: 10.3969/j.issn.1000-3428.2013.07.054

• 人工智能及识别技术 • 上一篇    下一篇

一种求解社区检测问题的改进分布估计算法

周本达1,王煦法2,姚宏亮3   

  1. (1. 皖西学院应用数学学院,安徽 六安 237012;2. 安徽省智能计算及其应用重点实验室,合肥 230037;3. 合肥工业大学计算机科学与技术系,合肥 230009)
  • 收稿日期:2012-07-05 出版日期:2013-07-15 发布日期:2013-07-12
  • 作者简介:周本达(1974-),男,副教授、硕士,主研方向:智能算法,复杂网络;王煦法,教授;姚宏亮,副教授、博士
  • 基金资助:
    国家自然科学基金资助项目(61175051, 61070131);安徽省高校省级自然科学研究基金资助重点项目(K2011A267, KJ2013 A259)

An Improved Estimation of Distribution Algorithm for Solving Community Detection Problem

ZHOU Ben-da 1, WANG Xu-fa 2, YAO Hong-liang 3   

  1. (1. School of Applied Mathematics, West Anhui University, Lu’an 237012, China; 2. Nature Inspired Computation and Its Applications Laboratory of Anhui Province, Hefei 230037, China; 3. School of Computer Science & Technology, Hefei University of Technology, Hefei 230009, China)
  • Received:2012-07-05 Online:2013-07-15 Published:2013-07-12

摘要: 在分析网络模块性函数局部单调性的基础上,设计局部搜索变异算子,提出一种求解社区检测问题的改进分布估计算法。基于基准测试网络和大规模复杂网络对算法进行测试,实验结果表明,对于不同网络,该算法运行100次得到的Q函数平均值均优于Girvan- Newman算法、Newman快速算法和Tasgin遗传算法。

关键词: 社区检测, 分布估计算法, 复杂网络, 局部搜索变异, 模块性函数

Abstract: Based on the analysis of local monotonic of modularity function, this paper designs a Local Search and Mutation(LSM) operator, and proposes an improved Estimation of Distribution Algorithm(EDA) for solving community detection problem. The proposed algorithm is tested on basic network and big scale complex network. Experimental results show that the Q function average values of this algorithm while running 100 in different networks times is better than Girvan-Newman(GN) algorithm, Fast Newman(FN) algorithm and Tasgin Genetic Algorithm(TGA).

Key words: community detection, Estimation of Distribution Algorithm(EDA), complex network, Local Search and Mutation(LSM), modular function

中图分类号: