Show simple item record

dc.contributor.authorMendialdua Beitia, Iñigo ORCID
dc.contributor.authorMartínez Otzeta, José María
dc.contributor.authorRodríguez Rodríguez, Igor ORCID
dc.contributor.authorRuiz Vázquez, María Consuelo
dc.contributor.authorSierra Araujo, Basilio ORCID
dc.date.accessioned2024-01-11T14:56:48Z
dc.date.available2024-01-11T14:56:48Z
dc.date.issued2015-05-19
dc.identifier.citationKnowledge-Based Systems 85 : 298-306 (2015)es_ES
dc.identifier.issn0950-7051
dc.identifier.issn1872-7409
dc.identifier.urihttp://hdl.handle.net/10810/63879
dc.description.abstractClass binarization strategies decompose the original multi-class problem into several binary sub-problems. One versus One (OVO) is one of the most popular class binarization techniques, which considers every pair of classes as a different sub-problem. Usually, the same classifier is applied to every sub-problem and then all the outputs are combined by some voting scheme. In this paper we present a novel idea where for each test instance we try to assign the best classifier in each sub-problem of OVO. To do so, we have used two simple Dynamic Classifier Selection (DCS) strategies that have not been yet used in this context. The two DCS strategies use K-NN to obtain the local region of the test-instance, and the classifier that performs the best for those instances in the local region, is selected to classify the new test instance. The difference between the two DCS strategies remains in the weight of the instance. In this paper we have also proposed a novel approach in those DCS strategies. We propose to use the K-Nearest Neighbor Equality (K-NNE) method to obtain the local accuracy. K-NNE is an extension of K-NN in which all the classes are treated independently: the K nearest neighbors belonging to each class are selected. In this way all the classes take part in the final decision. We have carried out an empirical study over several UCI databases, which shows the robustness of our proposal.es_ES
dc.description.sponsorshipThe work described in this paper was partially conducted within the Basque Government Research Team Grant IT313-10 and the University of the Basque Country UPV/EHU. I. Mendialdua holds a Grant from Basque Government.es_ES
dc.language.isoenges_ES
dc.publisherElsevieres_ES
dc.rightsinfo:eu-repo/semantics/openAccesses_ES
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.subjectmachine learninges_ES
dc.subjectsupervised classificationes_ES
dc.subjectdecomposition strategieses_ES
dc.subjectone against onees_ES
dc.subjectclassifier combinationes_ES
dc.subjectdynamic classifier selectiones_ES
dc.titleDynamic selection of the best base classifier in one versus onees_ES
dc.typeinfo:eu-repo/semantics/articlees_ES
dc.rights.holder© 2015 Elsevier under CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)es_ES
dc.relation.publisherversionhttps://www.sciencedirect.com/science/article/abs/pii/S0950705115002014es_ES
dc.identifier.doi10.1016/j.knosys.2015.05.015
dc.departamentoesArquitectura y Tecnología de Computadoreses_ES
dc.departamentoesCiencia de la computación e inteligencia artificiales_ES
dc.departamentoeuKonputagailuen Arkitektura eta Teknologiaes_ES
dc.departamentoeuKonputazio zientziak eta adimen artifizialaes_ES


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record

© 2015 Elsevier under CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Except where otherwise noted, this item's license is described as © 2015 Elsevier under CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)