Bridging the Gap between Naive Bayes and Maximum Entropy Text Classification

The naive Bayes and maximum entropy approaches to text classifi- cation are typically discussed as completely unrelated techniques. In this pa per, however, we show that both approaches are simply two different ways of doing parameter estimation for a common log-linear model of class posteriors. In par- ticular, we show how to map the solution given by maximum entropy into an optimal solution for naive Bayes according to the conditional maximum likeli- hood criterion.