Please use this identifier to cite or link to this item:
Files in This Item:
File Description SizeFormat 
renanpiazzarolifinottiamaral.pdf1.14 MBAdobe PDFThumbnail
Type: Dissertação
Title: Type-1 and singleton fuzzy logic system trained by a fast scaled conjugate gradient methods for dealing with classification problems
Author: Amaral, Renan Piazzaroli Finotti
First Advisor: Ribeiro, Moisés Vidal
Referee Member: Aguiar, Eduardo Pestana de
Referee Member: Silva Junior, Ivo Chaves da
Referee Member: Guimarães, Frederico Gadelha
Resumo: -
Abstract: This thesis presents and discusses improvements in the type-1 and singleton fuzzy logic system for dealing with classification problems. Two training methods are addressed, the scaled conjugate gradient, which uses the second order information approximating the multiplication of the Hessian matrix H by the directional vector v (i.e. Hv), and the same method using the differential operator R {.} to compute the exact value of Hv. Also, in order to adapt the fuzzy model to handle multiclass classification problems, it is developed a novel fuzzy model with a vector as output. All proposals are tested through the performance metrics analysis based on data sets provided by UCI Machine Learning Repository. The reported results show the high convergence speed and better classification rates of the proposed training methods than others presented in the literature. Additionally, the novel fuzzy model has a significant reduction in computational and classifier complexity, especially when the number of classes in classification problem increases.
Keywords: Fuzzy logic system
Multiclass classification
Scaled conjugate gradient
Language: eng
Country: Brasil
Publisher: Universidade Federal de Juiz de Fora (UFJF)
Institution Initials: UFJF
Department: ICE – Instituto de Ciências Exatas
Program: Programa de Pós-graduação em Engenharia Elétrica
Access Type: Acesso Aberto
Issue Date: 1-Sep-2017
Appears in Collections:Mestrado em Engenharia Elétrica (Dissertações)

Items in DSpace are protected by Creative Commons licenses, with all rights reserved, unless otherwise indicated.