The ’MXM’ (Mens eX Machina, meaning ’Mind from the Machine’ in Latin) package is an R package that implements feature selection methods for identifying minimal, statistically-equivalent and equally predictive feature subsets. Bayesian network algorithms and related functions are also included, while a variety of statistical conditional independence tests and cross validation strategies are available.

‘MXM’-package is released under the official public CRAN Repository of R. The current version of the package after several updates is 1.4.2.


Ioannis Tsamardinos, Vincenzo Lagani, Giorgos Athineou, Michail Tsagris, Giorgos Borboudakis & Anna Roumpelaki


2014-05-12 (Last update: 15/1/2019)

Reference manual



R (programming language)

R is a free software environment for statistical computing and graphics. It compiles and runs on a wide variety of UNIX platforms, Windows and MacOS


MXM package source


Installation of Packages in R: Please type help(“INSTALL”) or help(“install.packages”) in R for information on how to install packages from this repository. The manual R Installation and Administration (also contained in the R base sources) explains the process in detail.


For any question, suggestion or bug report please contact Michail Tsagris <>

Related Publications

Vincenzo Lagani, Giorgos Athineou, Alessio Farcomeni, Michail Tsagris, Ioannis Tsamardinos (2016): Feature Selection with the R Package MXM: Discovering Multiple, Statistically-Equivalent, Predictive Feature Subsets, Accepted for publication on Journal of Statistical Software