WEIGHTED NAÏVE BAYES FOR TEXT CLASSIFICATION USING POSITIVE TERM-CLASS DEPENDENCY
Yanjun Li, Congnan Luo, Soon Myoung Chung · International Journal of Artificial Intelligence Tools · 2011
Naïve Bayes is a simple and efficient classification algorithm which performs well on text classification, which is also known as text categorization. Many researches have been done to improve the performance of the naïve Bayes classifier by weighting the correlated terms, in order to relax the strong assumption of independence between terms. In this paper, we first introduce a new χ2 statistical data, denoted by Rw,c, which can measure positive term-class dependency accurately, and then propose a new weighted naïve Bayes classifier using Rw,c at the training phase. Experimental results with real data sets show that our weighted naïve Bayes classifier has much better performance than the basic naïve Bayes classifier in most cases.