Optimization of Classifiers Using Genetic Algorithms
Juan Julián Merelo, Abraham Prieto, Federico Morán · The MIT Press eBooks · 2001
this paper, we will try to present a general procedure for optimizing classifiers, based in a two-level GA operating on variable size chromosomes, that uses as fitness several of the classifier characteristics. We will first present the state of the art in evolutionary classifier design in section 2, then the methodology for genetic classifier design will be presented in section 3, to be followed by different implementations of the method to design unsupervised classifiers in section 4 and supervised classifiers in section 5.