Scaling Up the Accuracy of Naive Bayes Classi ers a Decision Tree Hybrid

Naive Bayes induction algorithms were previously shown to be surprisingly accurate on many classi cation tasks even when the conditional independence assumption on which they are based is violated How ever most studies were done on small databases We show that in some larger databases the accuracy of Naive Bayes does not scale up as well as decision trees We then propose a new algorithm NBTree which in duces a hybrid of decision tree classi ers and Naive Bayes classi ers the decision tree nodes contain uni variate splits as regular decision trees but the leaves contain Naive Bayesian classi ers The approach re tains the interpretability of Naive Bayes and decision trees while resulting in classi ers that frequently out perform both constituents especially in the larger databases tested