Performance analysis and improvement of naïve Bayes in text classification application

Naive Bayes classifier is widely used in machine learning for its simplicity and efficiency. However, most of the existing work on naïve Bayes focused on improving the Bayes model itself or whether the “naïve assumption” is satisfied. In this paper, the performance of naïve bayes in text classification is analyzed and the corresponding results from different points of view is proposed, then an improving way for text classification with highly asymmetric misclassification costs is provided. Finally the related experiments proved the above proposed method were efficient.