MXM: Discovering Multiple, Statistically-Equivalent Signatures

Feature selection methods for identifying minimal, statistically-equivalent and equally-predictive feature subsets. Bayesian network algorithms and related functions are also included. The package name 'MXM' stands for "Mens eX Machina", meaning "Mind from the Machine" in Latin.

Version: 0.9.9
Imports: methods, stats, utils, survival, MASS, graphics, ordinal, nnet, quantreg, lme4, foreach, doParallel, parallel, speedglm, e1071, relations, Rfast, visNetwork, energy
Suggests: hash, R.rsp
Published: 2017-03-26
Author: Ioannis Tsamardinos, Vincenzo Lagani, Giorgos Athineou, Michail Tsagris, Giorgos Borboudakis, Anna Roumpelaki
Maintainer: Michail Tsagris <mtsagris at>
License: GPL-2
NeedsCompilation: no
Citation: MXM citation info
In views: MachineLearning
CRAN checks: MXM results


Reference manual: MXM.pdf
Vignettes: R packages: LaTeX vignettes
Package source: MXM_0.9.9.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X El Capitan binaries: r-release: MXM_0.9.9.tgz
OS X Mavericks binaries: r-oldrel: MXM_0.9.9.tgz
Old sources: MXM archive


Please use the canonical form to link to this page.