【摘要】 社团结构是复杂网络的一种很普遍且非常重要的拓扑特征,社团的发现有助于了解复杂网络的结构和功能。节点间相似度的评价指标对于社团发现的结果起着至关重要的作用,传统算法中使用的相似度指标存在着时间复杂度过高和不够精确的缺陷。为了弥补这两个缺陷,在信息传递理论的基础上将网络中的节点抽象成了多维数据集,结合传统聚类算法K-means提出了一种社团发现的新算法。基于Zachary Karate Club网络、Jazz Musician网络和Facebook网络的实验结果表明,该算法是高效且准确的。更多还原

【Abstract】 Community structure is one of the ubiquitous and significant topology characteristics of complex network.It can help us to learn the structure and functions of a complex network.The similarity index plays a vital role in community detection but it has the shortage of high time complexity and low accuracy.In order to improve the two shortages,nodes are abstracted from a complex network into a multi-dimension data set based on the theory of information transmission in the network.Combined with the traditional clustering algorithm K-means,a new community detection algorithm was proposed.The experimental results obtained from Zachary Karate Club network,Jazz Musician network and Facebook network show that the algorithm is effective and accurate.更多还原