参考文档:http://blog.csdn.net/lawrencesgj/article/details/8606570 源代码: https://github.com/shenguojun/hadoop/tree/master/WebKmeans/src/edu/sysu/shen/hadoop https://github.com/shenguojun/hadoop/blob/master/WebKmeans/src/edu/sysu/shen/hadoop/Kmeans.java Canopy算法 http://blog.csdn.net/july_2/article/details/8905502