Fuzzy System Identification Using Gaussian Mixture Models

The thesis is structured into 4 major studies, each presented under a separate chapter.Chapter 1 - A theoretical bridge associating a Gaussian Mixture Model (GMM) to an additive fuzzy system is uncovered. Chapter 3 - Invoke the Jacobi vs Gauss_seidel analogy in an attempt to speed up the rate of EM...

Full description

Bibliographic Details
Main Author: Gan, Ming Tao
Format: Thesis
Published: 2007
Subjects:
Online Access:http://shdl.mmu.edu.my/1147/
Description
Summary:The thesis is structured into 4 major studies, each presented under a separate chapter.Chapter 1 - A theoretical bridge associating a Gaussian Mixture Model (GMM) to an additive fuzzy system is uncovered. Chapter 3 - Invoke the Jacobi vs Gauss_seidel analogy in an attempt to speed up the rate of EM convergence. Chapter 4 - The theses proposes an adaptation of the EM that minimizes the average predicted error, in conjunction with maximizing likelihood.