Text Categorization Based on LDA and SVM

Text categorization aims to assign text documents to predefined categories. In this paper, a novel text categorization algorithm that combines the LDA and SVM is proposed. The core idea of the algorithm is as follows: The high dimension text data set are first projected into a lower-dimensional text subspace. Then the SVM classifier algorithm is applied to classify the text. Experimental results on two text benchmark data sets demonstrate the effectiveness of the proposed text classification algorithm.