An Attribute Reduction Algorithm Based on Rough Set Theory and an Improved Genetic Algorithm

Abstract

Because the existing attribute reduction algorithms based on rough set theory and genetic algorithm have the main problems: the complexity in calculating fitness function and slow speed in convergence. An attribute reduction algorithm based on rough set theory and an improved genetic algorithm is proposed in this paper. In order to simplify the calculation of fitness function under the condition of keeping the algorithm correct, the relative importance of chromosome is used to define the fitness function. Beyond that, by introducing the core attributes into the initial population and using an improved mutation operator, the algorithm can not only maintain the feature of whole optimization, but also have a quicker convergence speed. The experimental results show that the algorithm can obtain the optimal attribute reduction set of information system quickly and efficiently.