1.Read a paper about document classification wiht GMM distributions of word vecotrs and try to code it in python 2.Use GibbsLDA++ to reduce the dimension of the text in 20news、r52、r8、webkb and contrast the performance of classification
