ebmc: Ensemble-Based Methods for Class Imbalance Problem

Four ensemble-based methods (SMOTEBoost, RUSBoost, UnderBagging, and SMOTEBagging) for class imbalance problem are implemented for binary classification. Such methods adopt ensemble methods and data re-sampling techniques to improve model performance in presence of class imbalance problem. One special feature offers the possibility to choose multiple supervised learning algorithms to build weak learners within ensemble models. References: Nitesh V. Chawla, Aleksandar Lazarevic, Lawrence O. Hall, and Kevin W. Bowyer (2003) <doi:10.1007/978-3-540-39804-2_12>, Chris Seiffert, Taghi M. Khoshgoftaar, Jason Van Hulse, and Amri Napolitano (2010) <doi:10.1109/TSMCA.2009.2029559>, R. Barandela, J. S. Sanchez, R. M. Valdovinos (2003) <doi:10.1007/s10044-003-0192-z>, Shuo Wang and Xin Yao (2009) <doi:10.1109/CIDM.2009.4938667>, Yoav Freund and Robert E. Schapire (1997) <doi:10.1006/jcss.1997.1504>.

Version: 1.0.0
Depends: methods
Imports: e1071, rpart, C50, randomForest, DMwR, pROC
Published: 2017-08-29
Author: Hsiang Hao, Chen
Maintainer: "Hsiang Hao, Chen" <kbman1101 at gmail.com>
License: GPL (≥ 3)
NeedsCompilation: no
CRAN checks: ebmc results


Reference manual: ebmc.pdf
Package source: ebmc_1.0.0.tar.gz
Windows binaries: r-devel: ebmc_1.0.0.zip, r-release: ebmc_1.0.0.zip, r-oldrel: ebmc_1.0.0.zip
OS X El Capitan binaries: r-release: ebmc_1.0.0.tgz
OS X Mavericks binaries: r-oldrel: ebmc_1.0.0.tgz


Please use the canonical form https://CRAN.R-project.org/package=ebmc to link to this page.