eprintid: 1027
rev_number: 6
eprint_status: archive
userid: 6
dir: disk0/00/00/10/27
datestamp: 2011-12-05 11:40:26
lastmod: 2011-12-05 11:40:26
status_changed: 2011-12-05 11:40:26
type: article
metadata_visibility: show
creators_name: Alexandridis, Alex
creators_name: Patrinos, Panagiotis
creators_name: Sarimveis, Haralambos
creators_name: Tsekouras, George
creators_id: 
creators_id: panagiotis.patrinos@imtlucca.it
creators_id: 
creators_id: 
title: A two-stage evolutionary algorithm for variable selection in the development of RBF neural network models
ispublished: pub
subjects: QA
subjects: QA75
divisions: CSA
full_text_status: none
keywords: Variable selection; Radial basis functions; Neural networks; Evolutionary computation; Genetic algorithms; Simulated annealing
abstract: In many modeling problems that are based on input–output data, information about a plethora of variables is available. In these cases, the proper selection of explanatory variables is very critical for the success of the produced model, since it eliminates noisy variables and possible correlations, reduces the size of the model and accomplishes more accurate predictions. Many variable selection procedures have been proposed in the literature, but most of them consider only linear models. In this work, we present a novel methodology for variable selection in nonlinear modeling, which combines the advantages of several artificial intelligence technologies. More specifically, the Radial Basis Function (RBF) neural network architecture serves as the nonlinear modeling tool, by exploiting the simplicity of its topology and the fast fuzzy means training algorithm. The proper variables are selected in two stages using a multi-objective optimization approach: in the first stage, a specially designed genetic algorithm minimizes the prediction error over a monitoring data set, while in the second stage a simulated annealing technique aims at the reduction of the number of explanatory variables. The efficiency of the proposed method is illustrated through its application to a number of benchmark problems.
date: 2005-02
date_type: published
publication: Chemometrics and Intelligent Laboratory Systems
volume: 75
number: 2
publisher: Elsevier
pagerange: 149-162
id_number: doi:10.1016/j.chemolab.2004.06.004
refereed: TRUE
issn: 0169-7439
official_url: http://dx.doi.org/10.1016/j.chemolab.2004.06.004
citation:   Alexandridis, Alex and Patrinos, Panagiotis and Sarimveis, Haralambos and Tsekouras, George  A two-stage evolutionary algorithm for variable selection in the development of RBF neural network models.  Chemometrics and Intelligent Laboratory Systems, 75 (2).  pp. 149-162.  ISSN 0169-7439      (2005)