Undirected cyclic graph based multiclass pair-wise classifier: Classifier number reduction maintaining accuracy

Supervised Classification approaches try to classify correctly the new unlabelled examples based on a set of well-labelled samples. Nevertheless, some classification methods were formulated for binary classification problems and has difficulties for multi-class problems. Binarization strategies deco...

Descripción completa

Detalles Bibliográficos
Autores: Mendialdua Beitia, Iñigo, Echegaray López, Goretti, Rodríguez Rodríguez, Igor, Lazkano Ortega, Elena, Sierra Araujo, Basilio
Tipo de recurso: artículo
Fecha de publicación:2015
País:España
Institución:Universidad del País Vasco
Repositorio:Addi. Archivo Digital para la Docencia y la Investigación
OAI Identifier:oai:addi.ehu.eus:10810/63880
Acceso en línea:http://hdl.handle.net/10810/63880
Access Level:acceso abierto
Palabra clave:machine learning
supervised classification
decomposition strategies
one-vs-one
id ES_f75a06f4d24cdb89e181fddb4de981b7
oai_identifier_str oai:addi.ehu.eus:10810/63880
network_acronym_str ES
network_name_str España
repository_id_str
spelling Undirected cyclic graph based multiclass pair-wise classifier: Classifier number reduction maintaining accuracyMendialdua Beitia, IñigoEchegaray López, GorettiRodríguez Rodríguez, IgorLazkano Ortega, ElenaSierra Araujo, Basiliomachine learningsupervised classificationdecomposition strategiesone-vs-oneSupervised Classification approaches try to classify correctly the new unlabelled examples based on a set of well-labelled samples. Nevertheless, some classification methods were formulated for binary classification problems and has difficulties for multi-class problems. Binarization strategies decompose the original multi-class dataset into multiple two-class subsets. For each new sub-problem a classifier is constructed. One-vs-One is a popular decomposition strategy that in each sub-problem discriminates the cases that belong to a pair of classes, ignoring the remaining ones. One of its drawbacks is that it creates a large number of classifiers, and some of them are irrelevant. In order to reduce the number of classifiers, in this paper we propose a new method called Decision Undirected Cyclic Graph. Instead of making the comparisons of all the pair of classes, each class is compared only with other two classes; evolutionary computation is used in the proposed approach in order to obtain suitable class pairing. In order to empirically show the performance of the proposed approach, a set of experiments over four popular Machine Learning algorithms are carried out, where our new method is compared with other well-known decomposition strategies of the literature obtaining promising results.The authors gratefully acknowledge J. Ceberio for his assistance during the work. The work described in this paper was partially conducted within the Basque Government Research Team Grant IT313-10. I. Mendialdua holds a Grant from Basque Government.Elsevier202420242015info:eu-repo/semantics/articleapplication/pdfhttp://hdl.handle.net/10810/63880reponame:Addi. Archivo Digital para la Docencia y la Investigacióninstname:Universidad del País VascoIngléshttps://www.sciencedirect.com/science/article/abs/pii/S0925231215010851info:eu-repo/semantics/openAccesshttp://creativecommons.org/licenses/by-nc-nd/4.0/© 2015 Elsevier B.V. under CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)oai:addi.ehu.eus:10810/638802026-06-18T09:23:17Z
dc.title.none.fl_str_mv Undirected cyclic graph based multiclass pair-wise classifier: Classifier number reduction maintaining accuracy
title Undirected cyclic graph based multiclass pair-wise classifier: Classifier number reduction maintaining accuracy
spellingShingle Undirected cyclic graph based multiclass pair-wise classifier: Classifier number reduction maintaining accuracy
Mendialdua Beitia, Iñigo
machine learning
supervised classification
decomposition strategies
one-vs-one
title_short Undirected cyclic graph based multiclass pair-wise classifier: Classifier number reduction maintaining accuracy
title_full Undirected cyclic graph based multiclass pair-wise classifier: Classifier number reduction maintaining accuracy
title_fullStr Undirected cyclic graph based multiclass pair-wise classifier: Classifier number reduction maintaining accuracy
title_full_unstemmed Undirected cyclic graph based multiclass pair-wise classifier: Classifier number reduction maintaining accuracy
title_sort Undirected cyclic graph based multiclass pair-wise classifier: Classifier number reduction maintaining accuracy
dc.creator.none.fl_str_mv Mendialdua Beitia, Iñigo
Echegaray López, Goretti
Rodríguez Rodríguez, Igor
Lazkano Ortega, Elena
Sierra Araujo, Basilio
author Mendialdua Beitia, Iñigo
author_facet Mendialdua Beitia, Iñigo
Echegaray López, Goretti
Rodríguez Rodríguez, Igor
Lazkano Ortega, Elena
Sierra Araujo, Basilio
author_role author
author2 Echegaray López, Goretti
Rodríguez Rodríguez, Igor
Lazkano Ortega, Elena
Sierra Araujo, Basilio
author2_role author
author
author
author
dc.subject.none.fl_str_mv machine learning
supervised classification
decomposition strategies
one-vs-one
topic machine learning
supervised classification
decomposition strategies
one-vs-one
description Supervised Classification approaches try to classify correctly the new unlabelled examples based on a set of well-labelled samples. Nevertheless, some classification methods were formulated for binary classification problems and has difficulties for multi-class problems. Binarization strategies decompose the original multi-class dataset into multiple two-class subsets. For each new sub-problem a classifier is constructed. One-vs-One is a popular decomposition strategy that in each sub-problem discriminates the cases that belong to a pair of classes, ignoring the remaining ones. One of its drawbacks is that it creates a large number of classifiers, and some of them are irrelevant. In order to reduce the number of classifiers, in this paper we propose a new method called Decision Undirected Cyclic Graph. Instead of making the comparisons of all the pair of classes, each class is compared only with other two classes; evolutionary computation is used in the proposed approach in order to obtain suitable class pairing. In order to empirically show the performance of the proposed approach, a set of experiments over four popular Machine Learning algorithms are carried out, where our new method is compared with other well-known decomposition strategies of the literature obtaining promising results.
publishDate 2015
dc.date.none.fl_str_mv 2015
2024
2024
dc.type.none.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv http://hdl.handle.net/10810/63880
url http://hdl.handle.net/10810/63880
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv https://www.sciencedirect.com/science/article/abs/pii/S0925231215010851
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
http://creativecommons.org/licenses/by-nc-nd/4.0/
© 2015 Elsevier B.V. under CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
eu_rights_str_mv openAccess
rights_invalid_str_mv http://creativecommons.org/licenses/by-nc-nd/4.0/
© 2015 Elsevier B.V. under CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:Addi. Archivo Digital para la Docencia y la Investigación
instname:Universidad del País Vasco
instname_str Universidad del País Vasco
reponame_str Addi. Archivo Digital para la Docencia y la Investigación
collection Addi. Archivo Digital para la Docencia y la Investigación
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869424874773020673
score 15,300719

Similares en LA Referencia