Genetic fuzzy clustering for the definition of fuzzy sets
Juan R. Velasco, Sergio López, Luis Magdalena · 2002
This paper presents a new algorithm for fuzzy clustering applied to the definition of fuzzy sets. The aim of this algorithm is to obtain a good fuzzy partition for a given variable. It will use a historic data file as input and uses genetic algorithms to evolve a population of fuzzy sets in order to obtain the best fuzzy partition. The main advantage of this algorithm is that it does not need previous knowledge on the number of fuzzy sets. This number is inferred by the algorithm itself. At the end of this paper, some results on real industrial data are presented.