I am afraid to see and study the text mining, the models I did not like will be not based on some distribution,
How to process the text mining , whether there exists some new approaches for text mining?