Rough set attribute reduction algorithm based on discrete differential evolution

In order to get minimal subsets of attributes,this paper proposes a rough set attribute reduction algorithm based on discrete differential evolution algorithm.Using a new range encoding mapping,the differential evolution algorithm can be used to deal with minimum attribute reduction problem.Secondly,a new calculation of the fitness function is proposed to control chromosomes'evolution in the direction of the minimum reduction.Experimental results show that the algorithm is effective,especially when data is large-scale,with fast convergence speed and more time saved in computation,which provides a new idea for attribute reduction.