基于稠密子团和边聚类系数的局部社团挖掘算法

发掘复杂网络的社团结构,有助于深入理解网络结构属性及其功能重要性。本文通过定义稠密子团,结合边聚类系数和局部模块度,提出一种DIDE社团挖掘算法。该算法通过选取稠密子团作为初始聚类团,利用边聚类系数扩张该稠密子团,最大化局部模块度值来生成社团结构。在计算机生成网络、三社团网络、Zachary网络和美国足球俱乐部网络上进行社团划分,验证该算法的可行性和有效性。

Detecting the community by integrating the dense sub-group and edge clustering coefficient

Detecting community structures in the complex networks can help us fully understanding the properties of networks structures and importance of their function. By defining the dense subgroup, combining edge clustering coefficient and local modularity, we proposed a DIDE algorithm for detecting community structures in this paper. We selected the dense sub group as initial cluster group, then expanded this dense subgroup using edge clustering coefficient, formed community structures by maximizing the value of local modularity. The simulation results on the computer generated network, the three groups network, Zachary network and American football club network show that DIDE is viable and effective.