Study on discretization in rough set based on genetic algorithm

Abstract

Discretization of attributes with real values in rough set is an important problem in data mining. It is different from the traditional discretization which has particular characteristic. Nguyen S. H. has given a detailed description about discretization in rough set. This paper gives a genetic algorithm aimed at discretization proposed by Nguyen S. H. And at the same time we make an experiment on several datasets from UCI Machine Learning Repository by this method. During the experiment, we constantly optimized genetic algorithm by adopting some optimization strategies. The experiment has proved that using genetic algorithm to solve discretization of rough set is efficient whatever in time complexity or in accuracy.