bootfs: Use multiple feature selection algorithms to derive robust feature sets for two or multiclass classification problems

To select a set of features used for successful classification of two or more groups of samples, multiple classification and feature selection algorithms are utilised. By combining the results of all methods and applying a bootstrapping approach a robust set of features with high power to distinguish the sample groups is selected.

Version: 1.4.2
Depends: pROC
Imports: igraph, ROCR, gbm, colorRamps, gplots, gtools, pamr, randomForest, Boruta, caret, tgp, mlegp, penalizedSVM
Suggests: parallel
Published: 2013-09-22
Author: Christian Bender
Maintainer: Christian Bender <christian.bender at>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: no
CRAN checks: bootfs results


Reference manual: bootfs.pdf
Vignettes: Bootstrapped Feature Selection - Introduction
Package source: bootfs_1.4.2.tar.gz
MacOS X binary: bootfs_1.4.2.tgz
Windows binary:
Old sources: bootfs archive