smotefamily: A Collection of Oversampling Techniques for Class Imbalance Problem Based on SMOTE

A collection of various oversampling techniques developed from SMOTE is provided. SMOTE is a oversampling technique which synthesizes a new minority instance between a pair of one minority instance and one of its K nearest neighbor. (see <> for more information) Other techniques adopt this concept with other criteria in order to generate balanced dataset for class imbalance problem.

Version: 1.1
Depends: R (≥ 3.0.0)
Suggests: FNN, dbscan, igraph
Published: 2017-08-11
Author: Wacharasak Siriseriwan [aut, cre]
Maintainer: Wacharasak Siriseriwan <wacharasak.s at>
License: GPL-2
NeedsCompilation: no
CRAN checks: smotefamily results


Reference manual: smotefamily.pdf
Package source: smotefamily_1.1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X El Capitan binaries: r-release: smotefamily_1.1.tgz
OS X Mavericks binaries: r-oldrel: smotefamily_1.1.tgz
Old sources: smotefamily archive


Please use the canonical form to link to this page.