Scheduling exploration/exploitation levels in genetically-generated fuzzy knowledge bases

In this paper we study the influence of the exploration/exploitation balance on the performances of a real binary/like coded genetic algorithm in automatically generating fuzzy knowledge bases from a set of numerical data. The influence is explored through different scheduling of crossover strategies throughout the evolution process. The aim is to prove the influence of a good balance between exploration and exploitation levels on the performances of the optimization algorithm, along with the influence of a good definition of the early versus late stages of the evolution.