TITLE:
Experiments with Two New Boosting Algorithms
AUTHORS:
Xiaowei Sun, Hongbo Zhou
KEYWORDS:
Boosting, Combination Method, TAN, BAN, Bayesian Network Classifier
JOURNAL NAME:
Intelligent Information Management,
Vol.2 No.6,
June
29,
2010
ABSTRACT: Boosting is an effective classifier combination method, which can improve classification performance of an unstable learning algorithm. But it dose not make much more improvement of a stable learning algorithm. In this paper, multiple TAN classifiers are combined by a combination method called Boosting-MultiTAN that is compared with the Boosting-BAN classifier which is boosting based on BAN combination. We describe experiments that carried out to assess how well the two algorithms perform on real learning problems. Fi- nally, experimental results show that the Boosting-BAN has higher classification accuracy on most data sets, but Boosting-MultiTAN has good effect on others. These results argue that boosting algorithm deserve more attention in machine learning and data mining communities.