use NLPIR and TF-IDF to get the text vecter , use hierarchical clustering to get the clusters.
1.需要Data文件夹，可以从上级目录copy
