一种有效的k-means聚类初始中心选取方法

针对传统k-means聚类算法中对初始聚类中心随意选取和人为指定的缺陷,提出一种改进的初始聚类中心的选取方法,利用差异矩阵将新的聚类初始中心计算方法用在传统的k-means算法思想中,对传统的k-means算法进行改进。降低k-means算法的复杂度和对异常点的敏感度,提高算法的可伸缩性。

An Effective Method for Initial Centrepoints of K-means Clustering

In this paper,a improved initial cluster center selection method is presented for the defects of the traditional k-means cluster algorithm on the initial cluster center selection,and uses the new cluster initial center calculation method on the mind of traditional k-means algorithm.It is shown that the new algorithm not only reduces the complexity of k-means algorithm,but also improves the scalability of the k-means algorithm.