Bagging is an ensemble learning technique. The name “Bagging” is an acronym derived from Bootstrap AGGregatING. Each member of the ensemble is constructed from a different training dataset. Each dataset is a bootstrap sample from the original. The models are combined by a uniform average or vote. Bagging works best with unstable learners, that is those that produce differing generalization patterns with small changes to the training data. Bagging therefore tends not to work well with linear models. See ensemble learning for more details.
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer Science+Business Media, LLC
About this entry
Cite this entry
(2011). Bagging. In: Sammut, C., Webb, G.I. (eds) Encyclopedia of Machine Learning. Springer, Boston, MA. https://doi.org/10.1007/978-0-387-30164-8_52
Download citation
DOI: https://doi.org/10.1007/978-0-387-30164-8_52
Publisher Name: Springer, Boston, MA
Print ISBN: 978-0-387-30768-8
Online ISBN: 978-0-387-30164-8
eBook Packages: Computer ScienceReference Module Computer Science and Engineering