提出了一个判别模糊聚类中聚类数有效性的新指标. 首先利用FCM算法对数据集进行模糊聚类,通过隶属度矩阵和聚类中心构建加权二分网络. 然后通过改进加权二分网络的模函数,定义一个新的聚类有效性指标. 为了检验该有效性指标的性能, 选取了三个常见的有效性指标在十五个数据集上进行了对比.实验结果表明, 该有效性指标具有较好的性能.

This paper proposes a novel validity index for fuzzy clustering. A weighted bipartite network can be constructed based on the matrix of membership degree and the centers of each cluster which are obtained by FCM algorithm. Then a novel validity index can be defined by improving the modularity function of bipartite network. In order to testify the performance of this new index, three popular validation indices are carried out on fifteen data sets. The comparison results show that the introduced index is superior to other three validity indices.