Avaliação estatística de métodos de poda aplicados em neurônios intermediários da rede neural MLP

dc.contributorUniversidade Estadual Paulista (UNESP)
dc.creatorSilvestre, Miriam Rodrigues
dc.creatorLing, Lee Luan
dc.date2014-05-27T11:22:03Z
dc.date2016-10-25T18:23:02Z
dc.date2014-05-27T11:22:03Z
dc.date2016-10-25T18:23:02Z
dc.date2006-12-01
dc.date.accessioned2017-04-06T01:21:47Z
dc.date.available2017-04-06T01:21:47Z
dc.identifierIEEE Latin America Transactions, v. 4, n. 4, p. 249-256, 2006.
dc.identifier1548-0992
dc.identifierhttp://hdl.handle.net/11449/69266
dc.identifierhttp://acervodigital.unesp.br/handle/11449/69266
dc.identifier10.1109/TLA.2006.4472121
dc.identifier2-s2.0-77958181102.pdf
dc.identifier2-s2.0-77958181102
dc.identifierhttp://dx.doi.org/10.1109/TLA.2006.4472121
dc.identifier.urihttp://repositorioslatinoamericanos.uchile.cl/handle/2250/890530
dc.descriptionThere are several papers on pruning methods in the artificial neural networks area. However, with rare exceptions, none of them presents an appropriate statistical evaluation of such methods. In this article, we proved statistically the ability of some methods to reduce the number of neurons of the hidden layer of a multilayer perceptron neural network (MLP), and to maintain the same landing of classification error of the initial net. They are evaluated seven pruning methods. The experimental investigation was accomplished on five groups of generated data and in two groups of real data. Three variables were accompanied in the study: apparent classification error rate in the test group (REA); number of hidden neurons, obtained after the application of the pruning method; and number of training/retraining epochs, to evaluate the computational effort. The non-parametric Friedman's test was used to do the statistical analysis.
dc.languagepor
dc.relationIEEE Latin America Transactions
dc.rightsinfo:eu-repo/semantics/openAccess
dc.subjectArtificial Neural Network
dc.subjectClassification error rate
dc.subjectClassification errors
dc.subjectComputational effort
dc.subjectExperimental investigations
dc.subjectHidden layers
dc.subjectHidden neurons
dc.subjectIntermedia
dc.subjectMLP neural networks
dc.subjectMultilayer perceptron neural networks
dc.subjectNon-parametric
dc.subjectPruning methods
dc.subjectStatistical analysis
dc.subjectStatistical evaluation
dc.subjectFunction evaluation
dc.subjectNeural networks
dc.titleAvaliação estatística de métodos de poda aplicados em neurônios intermediários da rede neural MLP
dc.typeOtro


Este ítem pertenece a la siguiente institución