CONSTRUCCIÓN DE RELACIONES DE SIMILARIDAD BORROSA BASADA EN LA MEDIDA CALIDAD DE LA SIMILARIDAD

Abstract

In this research it proposed used fuzzy relations to define a new metric called Measure Quality of Similarity. New metric used a binary fuzzy relation, which quantify the strength of the similarity relationship in a range of [0.1]. The fuzzy relation is characterized by a membership function, which in this case is defined by a similarity function based on Extended Rough Set theory. This research it also proposed a method called PSO+RST+FUZZY which combines the Measure Quality of Similarity and metaheuristic Particule Swarm Optimization (PSO) to calculate features weights; and its impact for improve the performance of the k-Nearest Neighbors. Experimental results in international database show an effective behavior for accuracy