Hybrid Genetic Algorithms and Clustering
Francisco Mateus Rocha Filho, Fernando A. C. Gomide · European Society for Fuzzy Logic and Technology Conference · 2005
This paper introduces a hybrid genetic algorithm that uses fuzzy c-means clustering technique as a mechanism to reduce fitness evaluations and to preserve solution quality. Population clustering provides a means to evaluate only the representative individual of each cluster instead of the whole population. The remaining individuals are indirectly evaluated. The aim is to maintain reasonable population size and to obtain near-optimal solutions. This is an important issue especially in large-scale, complex optimization and decision-making problems.