Abstract

Breiman's bagging and Freund and Schapire's boosting are recent methods for improving the predictive power of classifier learning systems. Both form a set of classifiers that are combined by voting, bagging by generating replicated bootstrap samples of the data, and boosting by adjusting the weights of training instances. This paper reports results of applying both techniques to a system that learns decision trees and testing on a representative collection of datasets. While both approaches substantially improve predictive accuracy, boosting shows the greater benefit. On the other hand, boosting also produces severe degradation on some datasets. A small change to the way that boosting combines the votes of learned classifiers reduces this downside and also leads to slightly better results on most of the datasets considered. Introduction Designers of empirical machine learning systems are concerned with such issues as the computational cost of the learning method and the accuracy and ...

Citations

4654	C4.5: Programs for Machine Learning - Quinlan - 1993
3618	Classification and regression trees - Breiman, Friedman, et al. - 1984
2313	Bagging predictors - Breiman - 1996
2101	A decision-theoretic generalization of on-line learning and an application to boosting - Freund, Schapire - 1997
763	Estimation of Dependencies Based on Empirical Data - Vapnik - 1979
533	Solving Multiclass Learning Problems via ErrorCorrecting Output Codes - Dietterich, Bakiri - 1995
189	Classi cation and Regression Trees - Breiman, Friedman, et al. - 1984
121	Learning classification trees - Buntine - 1992
105	A comparative evaluation of voting and meta-learning on partitioned data - Chan, Stolfo - 1995
81	Learning oblique decision trees - Heath, Kasif, et al. - 1993
62	Addressing the selective superiority problem: Automatic algorithm/model class selection - Brodley - 1993
59	Megainduction: a test flight - Catlett - 1991
52	Lookahead feature construction for learning hard concepts - Ragavan, Rendell - 1993
50	Inductive knowledge acquisition: a case study. Applications of expert systems - QUINLAN, COMPTON, et al. - 1987
46	Automatic parameter selection by minimizing estimated error - Kohavi, John - 1995
29	The problem of small disjuncts: Its remedy in decision trees - TING - 1994
29	An incremental method for finding multivariate splits for decision trees - Utgoff, Brodley - 1990
15	Learning classi cation trees - Buntine - 1993
14	Constructing nominal x-of-n attributes - Zheng - 1995
6	Megainduction: A test ight - Catlett - 1991
3	Experiments with a new boosting algorithm. Unpublished manuscript - Freund, Shapire - 1996
3	A comparativeevaluation of voting and meta-learning on partitioned data - Chan, Stolfo - 1995
3	An incremental method for nding multivariate splits for decision trees - E, Brodley - 1990