摘要
揭示复杂网络的社区结构,对于了解网络结构与分析网络特性有重要意义。将一个网络划分为几个不同的社区,其本质也就是在一定程度上最大化提取网络本身的主要信息,同时略去一些相对次要的信息。主成分分析(Principle Component Analysis,PCA)方法,正是一种从对象中提取主要信息,而忽略相对次要信息的多元统计分析方法。本文基于PCA的信息压缩思想,提出了一种分析复杂网络社区结构的新方法,并将其应用于分析空手道俱乐部网络(Zachary网络)、海豚网络(Lusseau网络)、政治书籍网络(Krebs网络)等网络的社区结构,并且与基于模块度矩阵的谱方法划分结果进行了比较,数值实验结果表明本文提出的方法是可行且有效的。
It is important to find the community structure in complex networks for m,derstanding the structure and characteristics of networks. The nature of partitioning a network into several communities is to extract the major information of the network to the maximal extent. The principle component analysis （PCA） method is just a multi-statistics analysis method which extracts the major information and ignores the less important information at the same time. In this article, based on PCA we propose a new method of analyzing the community structure in complex networks, and then analyze the karate club network （Zachary network）, the dolphin social network （Lusseau network） , and the political books network（Krebs network） by using this new method. Further more, we compare the computational results with modularity-based analysis methods. Computational results demonstrate that the proposed method is feasible and effective.
出处
《运筹与管理》
CSCD
2008年第6期144-149,共6页
Operations Research and Management Science
基金
国家自然科学基金资助项目(1057101870431001)
关键词
系统科学
复杂网络
社区结构
主成分分析
systems science
complex networks
community structure
principle component analysis
