一种基于动态网格技术的K-means初始质心选取算法

Anintial Centroid Selected Algorithm Based on Dynamic Grid Technology

摘要: 针对K-means算法随机选取中心点而无法得到全局最优误差平方和SSE的问题,提出了一种基于移动网格技术的K-means中心点选取算法,该算法在明显减少K-means算法的迭代次数的同时,近似得到SSE的全局最优.

Abstract: In order to solve the problem that K-means Algorithm cannot get global optimal SSE(sum of the squared error) by choosing centers randomly,this paper presents a new algorithm of choosing centers of K-means based on shifting grid,which decreases iterative time of K-means obviously and gets the global optimal of SSE approximately.