一种基于划分的聚类算法分析与改进

K-means 是一种基于划分的聚类算法,由于 K-means 算法在选择初始聚类中心时是随机选取 k 个点,因此一旦 k 个点选取不合理,将会误导聚类过程,得到一个不合理的聚类结果。在分析聚类结果对初值依赖性的基础上,对初值选取方法进行了分析和研究,采取“射靶”的原理进行类中心搜索。从实验结果中可以发现,改进后 K-means 得到的聚类结果更加稳定,对初始聚类中心的依赖性减弱了。

Analysis and improvement based on a partitional clustering algorithm

K-means is a partitioning algorithm,So it is obvious that outputs are especially sensitive to initial starting condition for random selections about K initial starting points,which will lead to bad solutions,so the quality of cluster relies on the initial starting points. This paper, analyses the method of random selection and proposes a method of serching initial starting points through aiming at target many times.It can demonstrate that the improved K-means algorithm can get better solution and is less sensitive to initial starting points.