Categorizing Text Documents Using Naïve Bayes, SVM and Logistic Regression

Abstract

Categorizing Text documents is the method of arranging different types of documents into labelled data. The field of this paper is to combine the Data mining Technology, Data extraction and Artificial Intelligence for text categorization. This paper will showcase the features of the technologies involved. There are three machine learning algorithms (SVM, Multinomial Naïve Bayes and Logistic Regression) used in this paper for text categorization, i.e. arrange documents into different categories of dataset 20 news groups. In the evaluation of the above classification techniques, SVM classifier outperforms other classifiers for text categorization.