Show simple item record  

dc.contributor.authorSun, Quan
dc.contributor.authorPfahringer, Bernhard
dc.date.accessioned2012-12-10T22:28:05Z
dc.date.available2012-12-10T22:28:05Z
dc.date.copyright2012
dc.date.issued2012
dc.identifier.citationSun, Q. & Pfahringer, B. (2012). Bagging ensemble selection for regression. In Lecture Notes in Computer Science 2012, Volume 7691, Al 2012: Advances in Artificial Intelligence. Pp 695-706.en_NZ
dc.identifier.isbn9783642351013
dc.identifier.urihttps://hdl.handle.net/10289/6940
dc.description.abstractBagging ensemble selection (BES) is a relatively new ensemble learning strategy. The strategy can be seen as an ensemble of the ensemble selection from libraries of models (ES) strategy. Previous experimental results on binary classification problems have shown that using random trees as base classifiers, BES-OOB (the most successful variant of BES) is competitive with (and in many cases, superior to) other ensemble learning strategies, for instance, the original ES algorithm, stacking with linear regression, random forests or boosting. Motivated by the promising results in classification, this paper examines the predictive performance of the BES-OOB strategy for regression problems. Our results show that the BES-OOB strategy outperforms Stochastic Gradient Boosting and Bagging when using regression trees as the base learners. Our results also suggest that the advantage of using a diverse model library becomes clear when the model library size is relatively large. We also present encouraging results indicating that the non negative least squares algorithm is a viable approach for pruning an ensemble of ensembles.en_NZ
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.publisherSpringeren_NZ
dc.rightsThis is the author's accepted version. The original publication is available at www.springerlink.com. Copyright Springer-Verlag Berlin Heidelberg 2012.en_NZ
dc.subjectMachine learning
dc.titleBagging ensemble selection for regressionen_NZ
dc.typeConference Contributionen_NZ
dc.identifier.doi10.1007/978-3-642-35101-3_59en_NZ


Files in this item

This item appears in the following Collection(s)

Show simple item record