A two-stage evolutionary algorithm for variable selection in the development of RBF neural network models

Alexandridis, A; Patrinos, P; Sarimveis, H; Tsekouras, G

dc.contributor.author	Alexandridis, A	en
dc.contributor.author	Patrinos, P	en
dc.contributor.author	Sarimveis, H	en
dc.contributor.author	Tsekouras, G	en
dc.date.accessioned	2014-03-01T01:21:47Z
dc.date.available	2014-03-01T01:21:47Z
dc.date.issued	2005	en
dc.identifier.issn	0169-7439	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/16373
dc.subject	Evolutionary computation	en
dc.subject	Genetic algorithms	en
dc.subject	Neural networks	en
dc.subject	Radial basis functions	en
dc.subject	Simulated annealing	en
dc.subject	Variable selection	en
dc.subject.classification	Automation & Control Systems	en
dc.subject.classification	Chemistry, Analytical	en
dc.subject.classification	Computer Science, Artificial Intelligence	en
dc.subject.classification	Instruments & Instrumentation	en
dc.subject.classification	Mathematics, Interdisciplinary Applications	en
dc.subject.classification	Statistics & Probability	en
dc.subject.other	accuracy	en
dc.subject.other	algorithm	en
dc.subject.other	analytical error	en
dc.subject.other	article	en
dc.subject.other	artificial intelligence	en
dc.subject.other	artificial neural network	en
dc.subject.other	bioavailability	en
dc.subject.other	chemometrics	en
dc.subject.other	data analysis	en
dc.subject.other	information processing	en
dc.subject.other	mathematical model	en
dc.subject.other	molecular evolution	en
dc.subject.other	particle size	en
dc.subject.other	prediction	en
dc.subject.other	priority journal	en
dc.subject.other	reaction optimization	en
dc.subject.other	reduction kinetics	en
dc.subject.other	simulation	en
dc.title	A two-stage evolutionary algorithm for variable selection in the development of RBF neural network models	en
heal.type	journalArticle	en
heal.identifier.primary	10.1016/j.chemolab.2004.06.004	en
heal.identifier.secondary	http://dx.doi.org/10.1016/j.chemolab.2004.06.004	en
heal.language	English	en
heal.publicationDate	2005	en
heal.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. (C) 2004 Elsevier B.V All rights reserved.	en
heal.publisher	ELSEVIER SCIENCE BV	en
heal.journalName	Chemometrics and Intelligent Laboratory Systems	en
dc.identifier.doi	10.1016/j.chemolab.2004.06.004	en
dc.identifier.isi	ISI:000227055000004	en
dc.identifier.volume	75	en
dc.identifier.issue	2	en
dc.identifier.spage	149	en
dc.identifier.epage	162	en