Árboles de Redes Neuronales Autoorganizativas

Abstract

Automatic pattern classification is a very important field of artificial intelligence. For this type of tasks, several techniques have been used. In this work, a combination of decision trees and self-organizing neural networks is presented as an alternative to attack the problem. For the construction of these trees growth processes were applied. In these processes, evaluation of the classification efficiency of one or several nodes in different configurations is necessary in order to make decisions to optimize the structure and performance of the self-organizing neural tree net. For this task, a group of coefficients that quantify the efficiency is defined and a growth algorithm based on these coefficients give an objective measure of the classification performance for simple structures. Finally, a comparison with other classification methods, using cross-validation methods and real and artificial databases, is carried out.