Title: Bagging different instead of similar models for regression and classification problems

Authors: Sotiris B. Kotsiantis, Dimitris N. Kanellopoulos

Addresses: Educational Software Development Laboratory, Department of Mathematics, University of Patras, University Campus, 26504, Rio, Patras, Greece. ' Educational Software Development Laboratory, Department of Mathematics, University of Patras, University Campus, 26504, Rio, Patras, Greece

Abstract: Even though many ensemble techniques have been proposed, there is as yet no clear picture of which method is best. In this study, we propose a technique that uses different subsets of the same training dataset with the concurrent usage of a voting (for classification problems) or averaging methodology (for regression problems) for combining different learners instead of similar learners. We performed a comparison of the proposed ensemble with other well known ensembles that use the same base learners and the proposed technique had better accuracy in most cases.

Keywords: classifiers; machine learning; data mining; regressors; classification; regression; voting; averaging; learner ensembles.

DOI: 10.1504/IJCAT.2010.030472

International Journal of Computer Applications in Technology, 2010 Vol.37 No.1, pp.20 - 28

Published online: 17 Dec 2009 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article