sar is a Rough Set-based Attribute Reduction (aka Feature Selection) implementation. This is an implementation of ideas described, among other places, in the following paper:

Qiang Shen and Alexios Chouchoulas, A Modular Approach to Generating Fuzzy Rules with Reduced Attributes for the Monitoring of Complex Systems. Engineering Applications of Artificial Intelligence, 13(3):263-278, 2000.

rsar reads in a MIMO (Multiple Input, Multiple Output) dataset, performs RS-based feature selection on it, and returns the selected feature subset.

Four versions of the QuickReduct algorithm are supported, QuickReduct, QuickReduct III, QuickReduct IV and QuickReduct V (progressively faster implementations). QuickReduct II is a backward elimination version of QuickReduct and is not supported yet neither is exhaustive search for reducts. -sar is a Rough Set-based Attribute Reduction (aka Feature Selection) implementation. This is an implementation of ideas described, among other places, in the following paper:

　 Qiang Shen and Alexios Chouchoulas, A Modular Approach to Generating Fuzzy Rules with Reduced Attributes for the Monitoring of Complex Systems. Engineering Applications of Artificial Intelligence, 13(3):263-278, 2000.

　rsar reads in a MIMO (Multiple Input, Multiple Output) dataset, performs RS-based feature selection on it, and returns the selected feature subset.

　 Four versions of the QuickReduct algorithm are supported, QuickReduct, QuickReduct III, QuickReduct IV and QuickReduct V (progressively faster implementations). QuickReduct II is a backward elimination version of QuickReduct and is not supported yet　neither is exhaustive search for reducts.