k-means 聚类过程演示

k-means是一种非监督 （从下图 0 当中我们可以看到训练数据并没有标签标注类别）的聚类算法：

K-Means clustering intends to partition n objects into k clusters in which each object belongs to the cluster with the nearest mean. This method produces exactly k different clusters of greatest possible distinction. The best number of clusters k leading to the greatest separation (distance) is not known as a priori and must be computed from the data. The objective of K-Means clustering is to minimize total intra-cluster variance, or, the squared error function: 

 

0.initial

 

1.select centroids randomly   

  

 

2.assign points

 

 3.update centroids

 4.reassign points

 5.update centroids

 

 6.reassign points

 

7.iteration

 

reference：

https://www.naftaliharris.com/blog/visualizing-k-means-clustering/

https://www.saedsayad.com/clustering_kmeans.htm

转载请注明来源：https://www.cnblogs.com/lnas01/p/10347650.html