Evolutive Improvement of Parameters in an Associative Classifier

This paper presents an effective method to improve some of the parameters in an associative classifier, thus increasing its performance. This is accomplished using the simplicity and symmetry of the differential evolution metaheuristic. When modifying some parameters contained in the Gamma associative classifier, which is a novel associative model for pattern classification, this model have been found to be more efficient in the correct discrimination of objects; experimental results show that applying evolutionary algorithms models the desired efficiency and robustness of the classifier model is achieved. In this first approach, improving the Gamma associative classifier is achieved by applying the differential evolution algorithm.