k-means 算法接受输入量 k ；然后将n个数据对象划分为 k个聚类以便使得所获得的聚类满足：同一聚类中的对象相似度较高；而不同聚类中的对象相似度较小。聚类相似度是利用各聚类中对象的均值所获得一个“中心对象”（引力中心）来进行计算的。-k-means algorithm accepts input k　then n data objects into k clusters in order to make clustering satisfy obtained: the objects in the same cluster high similarity　different clustering object similarity small. Cluster similarity is the use of the mean of each cluster of objects that get a " central object"　(center of gravity) to perform the calculation.