regfilter: Elimination of Noisy Samples in Regression Datasets using Noise Filters

Traditional noise filtering methods aim at removing noisy samples from a classification dataset. This package adapts classic and recent filtering techniques to be used in regression problems. To do this, it uses the approach proposed in Martin (2021) [<doi:10.1109/ACCESS.2021.3123151>]. Thus, the goal of the implemented noise filters is to eliminate samples with noise in regression datasets.

Version: 1.0.3
Depends: R (≥ 3.2.0)
Imports: e1071, FNN, gbm, modelr, nnet, randomForest, rpart
Suggests: testthat (≥ 3.0.0), knitr, rmarkdown
Published: 2023-04-27
Author: Juan Martin [aut, cre], José A. Sáez [aut], Emilio Corchado [aut], Pablo Morales [ctb] (Author of the NoiseFiltersR package), Julian Luengo [ctb] (Author of the NoiseFiltersR package), Luis P.F. Garcia [ctb] (Author of the NoiseFiltersR package), Ana C. Lorena [ctb] (Author of the NoiseFiltersR package), Andre C.P.L.F. de Carvalho [ctb] (Author of the NoiseFiltersR package), Francisco Herrera [ctb] (Author of the NoiseFiltersR package)
Maintainer: Juan Martin <juanmartin at>
License: GPL (≥ 3)
Copyright: see file COPYRIGHTS
NeedsCompilation: no
Materials: NEWS
CRAN checks: regfilter results


Reference manual: regfilter.pdf
Vignettes: regfilter


Package source: regfilter_1.0.3.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): regfilter_1.0.3.tgz, r-oldrel (arm64): regfilter_1.0.3.tgz, r-release (x86_64): regfilter_1.0.3.tgz, r-oldrel (x86_64): regfilter_1.0.3.tgz
Old sources: regfilter archive


Please use the canonical form to link to this page.