聚类算法是给一大堆原始数据，然后通过算法将其中具有相似特征的数据聚为一类。

这里的k-means聚类，是事先给出原始数据所含的类数，然后将含有相似特征的数据聚为一个类中。(The clustering algorithm is given to a large number of original data, and then the data with similar features are gathered into a class by algorithm.
The K-means clustering here is to give the number of classes contained in the original data in advance, and then gather the data with similar features into a class.)