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

dc.contributor.author	Αλεξανδρίδης, Αλέξανδρος Π.	el
dc.contributor.author	Πατρινός, Παναγιώτης Κ.	el
dc.contributor.author	Σαρίμβεης, Χαράλαμπος Κ.	el
dc.contributor.author	Τσεκούρας, Γεώργιος Ε.	el
dc.date.accessioned	2015-06-04T13:30:47Z
dc.date.available	2015-06-04T13:30:47Z
dc.date.issued	2015-06-04
dc.identifier.uri	http://hdl.handle.net/11400/15060
dc.rights	Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ηνωμένες Πολιτείες	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/3.0/us/	*
dc.source	http://www.elsevier.com/	en
dc.subject	Evolutionary computation
dc.subject	Genetic algorithms
dc.subject	Neural networks
dc.subject	Radial basis functions
dc.subject	Simulated annealing
dc.subject	Variable selection
dc.subject	Εξελικτική υπολογιστική
dc.subject	Γενετικοί αλγόριθμοι
dc.subject	Νευρωνικά δίκτυα
dc.subject	Ακτινική συνάρτηση βάσης
dc.subject	Προσομοιωμένη ανόπτηση
dc.subject	Επιλογή μεταβλητής
dc.title	A two-stage evolutionary algorithm for variable selection in the development of RBF neural network models	en
heal.type	journalArticle
heal.classification	Technology
heal.classification	Electrical engineering
heal.classification	Τεχνολογία
heal.classification	Ηλεκτρολογία Μηχανολογία
heal.classificationURI	http://id.loc.gov/authorities/subjects/sh85133147
heal.classificationURI	http://zbw.eu/stw/descriptor/18426-4
heal.classificationURI	N/A-Τεχνολογία
heal.classificationURI	N/A-Ηλεκτρολογία Μηχανολογία
heal.keywordURI	http://id.loc.gov/authorities/subjects/sh95003989
heal.keywordURI	http://id.loc.gov/authorities/subjects/sh92002377
heal.keywordURI	http://zbw.eu/stw/descriptor/19808-6
heal.keywordURI	http://id.loc.gov/authorities/subjects/sh2002004691
heal.identifier.secondary	DOI: 10.1016/j.chemolab.2004.06.004
heal.language	en
heal.access	campus
heal.publicationDate	2005-02-28
heal.bibliographicCitation	ALEXANDRIDIS, A.P., PATRINOS, P.K., SARIMVEIS, H.K. & TSEKOURAS, G.E. (2005). A two-stage evolutionary algorithm for variable selection in the development of RBF neural network models. Chemometrics and Intelligent Laboratory Systems. [online] 75 (2). p. 149-162. Available from: http://www.elsevier.com/[Accessed 17/08/2004]	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.	en
heal.publisher	Elsevier	en
heal.journalName	Chemometrics and Intelligent Laboratory Systems	en
heal.journalType	peer-reviewed
heal.fullTextAvailability	true