Boosting naive bayes by active learning

AdaBoost has been proved to be an effective method to improve the performance of base classifiers both theoretically and empirically. However, previous studies have shown that AdaBoost can not obviously improve the performance of Naive Bayes as expected. This paper presents a new boosting algorithm, ActiveBoost, which applies active learning to mitigate the negative effect of noise data and introduce instability into boosting procedure. Empirical studies on a set of natural domains show that ActiveBoost has clear advantages with respect to the increasing of the classification accuracy of Naive Bayes when compared against Adaboost.