【摘要】 模糊C均值聚类(FCM)算法是一种经典的模糊聚类分析方法,但其算法初始聚类中心集是随机选取的,从而造成算法的性能强烈的依赖聚类中心集的初始化。提出了一种改进的基于多项式求解的FCM(PFCM)算法,该算法基于求解多项式的根来确定数据集初始聚类中心集,很好地解决了数据初始聚类中心集问题,使数据初始聚类中心集代表了数据集类别的特征,在此基础上,采用FCM算法得到聚类中心集的近似最优解。更多还原

【Abstract】 Fuzzy C-Means(FCM) algorithm is one of the most popular methods of clustering analysis. However, the traditional FCM algorithm does not work well because its initial clustering central collection is the stochastic selection. An efficient PFCM algorithm was proposed. Based on the solving multinomial root, the PFCM algorithm solved question of initial clustering central collection of data set. The experiment result demonstrates its effectiveness.更多还原