Text tendency categorization method based on class space model

Abstract

Text tendency categorization has important application in the content security management and public opinion analysis. Based on the studies of existing classification methods and the complexity of text tendency categorization, a text tendency categorization method based on class space model was proposed. By means of combined feature selection method, the new method realized text tendency classification based on the words of the types of tendency. Experimental results show that the method is effective in improving the accuracy and speed of text tendency categorization.