hyperSMURF: Hyper-Ensemble Smote Undersampled Random Forests

Machine learning supervised method to learn rare genomic features in imbalanced genetic data sets. This method can be also applied to classify or rank examples characterized by a high imbalance between the minority and majority class. hyperSMURF adopts a hyper-ensemble (ensemble of ensembles) approach, undersampling of the majority class and oversampling of the minority class to learn highly imbalanced data. Both single-core and parallel multi-core version of hyperSMURF are implemented.

Version: 1.1.2
Imports: unbalanced, randomForest, foreach, iterators, doParallel, parallel
Published: 2016-08-18
Author: Giorgio Valentini [aut, cre] - AnacletoLab, Dipartimento di Informatica, Universita' degli Studi di Milano; Max Schubach [ctb] - Charite, Universitatsmedizin Berlin; Matteo Re [ctb] - AnacletoLab, Dipartimento di Informatica, Universita' degli Studi di Milano; Peter N Robinson [ctb] - The Jackson Laboratory for Genomic Medicine, Farmington CT, USA.
Maintainer: Giorgio Valentini <valentini at di.unimi.it>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: no
CRAN checks: hyperSMURF results

Downloads:

Reference manual: hyperSMURF.pdf
Package source: hyperSMURF_1.1.2.tar.gz
Windows binaries: r-devel: hyperSMURF_1.1.2.zip, r-release: hyperSMURF_1.1.2.zip, r-oldrel: hyperSMURF_1.1.2.zip
OS X El Capitan binaries: r-release: hyperSMURF_1.1.2.tgz
OS X Mavericks binaries: r-oldrel: hyperSMURF_1.1.2.tgz
Old sources: hyperSMURF archive

Linking:

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