A parallel genetic algorithm based on linkage identification

Abstract

Linkage identification algorithms identify linkage groups - sets of loci tightly linked - before genetic optimizations for their recombination operators to work effectively and reliably. This paper proposes a parallel genetic algorithm (GA) based on the linkage identification algorithm and shows its effectiveness compared with other conventional parallel GAs such as master-slave and island models. This paper also discusses applicability of the parallel GAs that tries to answer "which method of the parallel GA should be employed to solve a problem?" © Springer-Verlag Berlin Heidelberg 2003.