K-means算法是硬聚类算法，是典型的基于原型的目标函数聚类方法的代表，它是数据点到原型的某种距离作为优化的目标函数，利用函数求极值的方法得到迭代运算的调整规则。K-means算法以欧式距离作为相似度测度，它是求对应某一初始聚类中心向量V最优分类，使得评价指标J最小。算法采用误差平方和准则函数作为聚类准则函数。(The K-means algorithm is a hard clustering algorithm, which is representative of the prototype based objective function clustering method. It is the distance from the data point to the prototype as the objective function of the optimization, and the method of using the function to find the extremum is used to get the adjustment rules of the iterative operation. The K-means algorithm takes Euclidean distance as the similarity measure, it is to find the V optimal classification corresponding to an initial cluster center vector, so that the evaluation index J is the smallest. The error square sum criterion function is used as a clustering criterion function.)