A technique for improving the performance of naive bayes text classification

Abstract

Naive Bayes classifier is widely used in text classification tasks, and it can perform surprisingly well, it is often regarded as a baseline. But previous researches show that the skewed distribution of training collection may cause poor results in text classification. This paper presents a new method to deal with this situation. We introduce a conditional probability which takes into account both the information of the whole corpus and each category. Our proposed method performs well in the standard benchmark collections, competing with the state-of-the-art text classifiers especially for the skewed data. © 2011 Springer-Verlag.