<!-- Title: R.ROSETTA Description: ROSETTA rough set classifier as a R package. Author: Mateusz Garbulowski --> <meta name='keywords' content='rosetta, rough sets, classification, transparent machine learning'>

R.ROSETTA

R.ROSETTA is an R package for constructing and analyzing rule-based classification models. R.ROSETTA is designed to support the overall data mining and knowledge discovery process. The presented tool is a more accessible and extended version of ROSETTA system (Øhrn, A., Komorowski, J., Skowron, A., & Synak, P., 1998). In addition to all the existing ROSETTA functionalities, we have added new functions such as: balancing data with undersampling, estimating rule P values, retrieving support sets from rules, mergind rule-based models, predicting external classes, visualizing rule-based model and generating synthetic data.

For more information and tutorials, please visit the official R.ROSETTA website.

Prerequisites

R.ROSETTA works with UNIX and Windows OS. However, UNIX operating systems (like MAC or Linux) require 32-bit Wine - a free and open-source compatibility layer. Please notice that latest version of macOS (Catalina) no longer supports 32-bit apps. Thus, we suggest to use VirtualBox or Docker.

Installation

Installation from github requires devtools package:

install.packages("devtools")

Installation and loading R.ROSETTA package from github:

library(devtools)
install_github("komorowskilab/R.ROSETTA")

library(R.ROSETTA)

Acknowledgments

R.ROSETTA includes a sample dataset collected from GEO repository with the reference number GSE25507.

Built With

• ROSETTA - The ROSETTA framework
• R Core - R Core Team

Related work

• ROSETTA - ROSETTA Technical Reference Manual
• VisuNet - VisuNet: an interactive tool for rule network visualization of rule-based machine learning models

Citation

Garbulowski M, Diamanti K, Smolińska K, Baltzer N, Stoll P, Bornelöv S, Øhrn A, Feuk L, Komorowski J. R.ROSETTA: an interpretable machine learning framework. BMC Bioinformatics. 2021 Mar 6;22(1):110. doi: 10.1186/s12859-021-04049-z link