ROSE: ROSE: Random Over-Sampling Examples

Share:

The package provides functions to deal with binary classification problems in the presence of imbalanced classes. Synthetic balanced samples are generated according to ROSE (Menardi and Torelli, 2013). Functions that implement more traditional remedies to the class imbalance are also provided, as well as different metrics to evaluate a learner accuracy. These are estimated by holdout, bootstrap or cross-validation methods.

Author
Nicola Lunardon, Giovanna Menardi, Nicola Torelli
Date of publication
2014-07-15 18:41:03
Maintainer
Nicola Lunardon <lunardon@stat.unipd.it>
License
GPL-2
Version
0.0-3

View on CRAN

Man pages

accuracy.meas
Metrics to evaluate a classifier accuracy in imbalanced...
hacide
Half circle filled data
ovun.sample
Over-sampling, under-sampling, combination of over- and...
roc.ROSE
ROC curve
ROSE
Generation of synthetic data by Randomly Over Sampling...
ROSE.eval
Evaluation of learner accuracy by ROSE
ROSE-package
ROSE: Random Over-Sampling Examples

Files in this package

ROSE
ROSE/inst
ROSE/inst/CITATION
ROSE/inst/ChangeLog
ROSE/NAMESPACE
ROSE/data
ROSE/data/hacide.rda
ROSE/R
ROSE/R/ROSE_eval.R
ROSE/R/estimation_funcs.R
ROSE/R/data_balancing_funcs.R
ROSE/MD5
ROSE/DESCRIPTION
ROSE/man
ROSE/man/ovun.sample.Rd
ROSE/man/hacide.Rd
ROSE/man/ROSE.eval.Rd
ROSE/man/roc.ROSE.Rd
ROSE/man/ROSE-package.Rd
ROSE/man/ROSE.Rd
ROSE/man/accuracy.meas.Rd