一种改进的K—means聚类算法

K—means算法是最常用的一种基于划分的聚类算法，但该算法需要事先指定K值、随机选择初始聚类中心等的缺陷，从而影响了K—means聚类结果的稳定性。针对K—means算法中的初始聚类中心是随机选择这一缺点进行改进，利用提出的新算法确定初始聚类中心，然后进行聚类，得出最终的聚类结果。实验证明，该改进算法比随机选择初始聚类中心的算法性能得到了提高，并且具有更高的准确性及稳定性。

An improved K-means clustering algorithm

K-means algorithm is one of the most commonly used clustering algorithm. But in actual application, there are some defects, for example, the value of K need to be specified ahead, and initial clustering center is a random choice and so on. This influences the performance of the K-menas algorithm. Aiming at the defect that the initial algorithm center of K-means is a random choice, this essay gives an improvement algorithm. Using this improved algorithm to comfirm clustering center to do clustering. After analysis, this improved algorithm makes the performance and accuracy better than the algorithm that random selection of initial clustering center.