Design of Multidimensional Classifiers using Fuzzy Brain Emotional Learning Model and Particle Swarm Optimization Algorithm

This study presents a multidimensional classifier design using a fuzzy brain emotional learning model, combined with a particle swarm optimization (PSO) algorithm that allows a network to automatically determine the optimum values for the weights of the reward signal. The multidimensional fuzzy brain emotional learning classifier (MFBELC) is first described with corresponding fuzzy inference rules; then the PSO algorithm is applied for the optimum parameter choice. This PSO-MFBELC is evaluated for the Wine dataset and Iris dataset, which are publicly available from the UCI machine learning database. A comparison of simulations using the proposed PSO-MFBELC shows that this classifier is superior to other algorithms in the recognition accuracy aspect.