Rough set based feature selection: A Review

Rough set is a tool with a mathematical foundation to deal with imprecise and imperfect knowledge. It has been widely applied in machine learning, data mining and knowledge discovery. One of the applications of Rough set theory in machine learning is the so-called feature selection especially for classification problems. This is performed by means of finding a reduct set of attributes. Reduct set is a subset of all features which retains classification accuracy as original attributes. Finding a reduct set in decision systems is NP-hard problem which has attracted many researchers to combine different methods with rough set. This paper is a survey of several methods of feature selection using rough set theory.