基于转移矩阵P的复杂网络聚类方法

A CLUSTERING ALGORITHM OF COMPLEX NETWORK BASED ON TRANSITION MATRIX P

提出了一种相对简单、有效的划分复杂网络社团结构的方法.该算法利用复杂网络的转移矩阵P和K均值聚类算法来划分社团结构,并且用F统计量判定最优的聚类结果,在探测社团结构明显的人工网时具有较高的准确度.

A relatively simple and effective clustering algorithm of complex network is proposed.This algorithm addresses community structure based on transition matrix P and K-means clustering.F statistics is used to give the best cluster number.It has high accuracy when detecting community structure of artificial networks with obvious structure.