Solving nonstationary classification problems with coupled support vector machines

Autores
Grinblat, Guillermo Luis; Uzal, Lucas César; Ceccatto, Hermenegildo A.; Granitto, Pablo Miguel
Año de publicación
2011
Idioma
inglés
Tipo de recurso
artículo
Estado
versión publicada
Descripción
Many learning problems may vary slowly over time: in particular, some critical real-world applications. When facing this problem, it is desirable that the learning method could find the correct input-output function and also detect the change in the concept and adapt to it. We introduce the time-adaptive support vector machine (TA-SVM), which is a new method for generating adaptive classifiers, capable of learning concepts that change with time. The basic idea of TA-SVM is to use a sequence of classifiers, each one appropriate for a small time window but, in contrast to other proposals, learning all the hyperplanes in a global way. We show that the addition of a new term in the cost function of the set of SVMs (that penalizes the diversity between consecutive classifiers) produces a coupling of the sequence that allows TA-SVM to learn as a single adaptive classifier. We evaluate different aspects of the method using appropriate drifting problems. In particular, we analyze the regularizing effect of changing the number of classifiers in the sequence or adapting the strength of the coupling. A comparison with other methods in several problems, including the well-known STAGGER dataset and the real-world electricity pricing domain, shows the good performance of TA-SVM in all tested situations.
Fil: Grinblat, Guillermo Luis. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; Argentina
Fil: Uzal, Lucas César. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; Argentina
Fil: Ceccatto, Hermenegildo A.. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; Argentina
Fil: Granitto, Pablo Miguel. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; Argentina
Materia
Adaptive Methods
Drifting Concepts
Support Vector Machine
Nivel de accesibilidad
acceso abierto
Condiciones de uso
https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
Repositorio
CONICET Digital (CONICET)
Institución
Consejo Nacional de Investigaciones Científicas y Técnicas
OAI Identificador
oai:ri.conicet.gov.ar:11336/15248

id CONICETDig_3581aeb0d87784700ef7365556b5e94d
oai_identifier_str oai:ri.conicet.gov.ar:11336/15248
network_acronym_str CONICETDig
repository_id_str 3498
network_name_str CONICET Digital (CONICET)
spelling Solving nonstationary classification problems with coupled support vector machinesGrinblat, Guillermo LuisUzal, Lucas CésarCeccatto, Hermenegildo A.Granitto, Pablo MiguelAdaptive MethodsDrifting ConceptsSupport Vector Machinehttps://purl.org/becyt/ford/1.2https://purl.org/becyt/ford/1Many learning problems may vary slowly over time: in particular, some critical real-world applications. When facing this problem, it is desirable that the learning method could find the correct input-output function and also detect the change in the concept and adapt to it. We introduce the time-adaptive support vector machine (TA-SVM), which is a new method for generating adaptive classifiers, capable of learning concepts that change with time. The basic idea of TA-SVM is to use a sequence of classifiers, each one appropriate for a small time window but, in contrast to other proposals, learning all the hyperplanes in a global way. We show that the addition of a new term in the cost function of the set of SVMs (that penalizes the diversity between consecutive classifiers) produces a coupling of the sequence that allows TA-SVM to learn as a single adaptive classifier. We evaluate different aspects of the method using appropriate drifting problems. In particular, we analyze the regularizing effect of changing the number of classifiers in the sequence or adapting the strength of the coupling. A comparison with other methods in several problems, including the well-known STAGGER dataset and the real-world electricity pricing domain, shows the good performance of TA-SVM in all tested situations.Fil: Grinblat, Guillermo Luis. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; ArgentinaFil: Uzal, Lucas César. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; ArgentinaFil: Ceccatto, Hermenegildo A.. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; ArgentinaFil: Granitto, Pablo Miguel. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; ArgentinaInstitute Of Electrical And Electronics Engineers2011-01info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_6501info:ar-repo/semantics/articuloapplication/pdfapplication/pdfapplication/pdfapplication/pdfhttp://hdl.handle.net/11336/15248Grinblat, Guillermo Luis; Uzal, Lucas César; Ceccatto, Hermenegildo A.; Granitto, Pablo Miguel; Solving nonstationary classification problems with coupled support vector machines; Institute Of Electrical And Electronics Engineers; Ieee Transactions On Neural Networks; 22; 1; 1-2011; 37-511045-92271941-0093enginfo:eu-repo/semantics/altIdentifier/doi/10.1109/TNN.2010.2083684info:eu-repo/semantics/altIdentifier/url/http://ieeexplore.ieee.org/document/5624639/?tp=&arnumber=5624639info:eu-repo/semantics/openAccesshttps://creativecommons.org/licenses/by-nc-sa/2.5/ar/reponame:CONICET Digital (CONICET)instname:Consejo Nacional de Investigaciones Científicas y Técnicas2025-10-15T14:55:51Zoai:ri.conicet.gov.ar:11336/15248instacron:CONICETInstitucionalhttp://ri.conicet.gov.ar/Organismo científico-tecnológicoNo correspondehttp://ri.conicet.gov.ar/oai/requestdasensio@conicet.gov.ar; lcarlino@conicet.gov.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:34982025-10-15 14:55:52.144CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicasfalse
dc.title.none.fl_str_mv Solving nonstationary classification problems with coupled support vector machines
title Solving nonstationary classification problems with coupled support vector machines
spellingShingle Solving nonstationary classification problems with coupled support vector machines
Grinblat, Guillermo Luis
Adaptive Methods
Drifting Concepts
Support Vector Machine
title_short Solving nonstationary classification problems with coupled support vector machines
title_full Solving nonstationary classification problems with coupled support vector machines
title_fullStr Solving nonstationary classification problems with coupled support vector machines
title_full_unstemmed Solving nonstationary classification problems with coupled support vector machines
title_sort Solving nonstationary classification problems with coupled support vector machines
dc.creator.none.fl_str_mv Grinblat, Guillermo Luis
Uzal, Lucas César
Ceccatto, Hermenegildo A.
Granitto, Pablo Miguel
author Grinblat, Guillermo Luis
author_facet Grinblat, Guillermo Luis
Uzal, Lucas César
Ceccatto, Hermenegildo A.
Granitto, Pablo Miguel
author_role author
author2 Uzal, Lucas César
Ceccatto, Hermenegildo A.
Granitto, Pablo Miguel
author2_role author
author
author
dc.subject.none.fl_str_mv Adaptive Methods
Drifting Concepts
Support Vector Machine
topic Adaptive Methods
Drifting Concepts
Support Vector Machine
purl_subject.fl_str_mv https://purl.org/becyt/ford/1.2
https://purl.org/becyt/ford/1
dc.description.none.fl_txt_mv Many learning problems may vary slowly over time: in particular, some critical real-world applications. When facing this problem, it is desirable that the learning method could find the correct input-output function and also detect the change in the concept and adapt to it. We introduce the time-adaptive support vector machine (TA-SVM), which is a new method for generating adaptive classifiers, capable of learning concepts that change with time. The basic idea of TA-SVM is to use a sequence of classifiers, each one appropriate for a small time window but, in contrast to other proposals, learning all the hyperplanes in a global way. We show that the addition of a new term in the cost function of the set of SVMs (that penalizes the diversity between consecutive classifiers) produces a coupling of the sequence that allows TA-SVM to learn as a single adaptive classifier. We evaluate different aspects of the method using appropriate drifting problems. In particular, we analyze the regularizing effect of changing the number of classifiers in the sequence or adapting the strength of the coupling. A comparison with other methods in several problems, including the well-known STAGGER dataset and the real-world electricity pricing domain, shows the good performance of TA-SVM in all tested situations.
Fil: Grinblat, Guillermo Luis. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; Argentina
Fil: Uzal, Lucas César. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; Argentina
Fil: Ceccatto, Hermenegildo A.. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; Argentina
Fil: Granitto, Pablo Miguel. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Rosario. Centro Internacional Franco Argentino de Ciencias de la Información y Sistemas; Argentina. Universidad Nacional de Rosario; Argentina
description Many learning problems may vary slowly over time: in particular, some critical real-world applications. When facing this problem, it is desirable that the learning method could find the correct input-output function and also detect the change in the concept and adapt to it. We introduce the time-adaptive support vector machine (TA-SVM), which is a new method for generating adaptive classifiers, capable of learning concepts that change with time. The basic idea of TA-SVM is to use a sequence of classifiers, each one appropriate for a small time window but, in contrast to other proposals, learning all the hyperplanes in a global way. We show that the addition of a new term in the cost function of the set of SVMs (that penalizes the diversity between consecutive classifiers) produces a coupling of the sequence that allows TA-SVM to learn as a single adaptive classifier. We evaluate different aspects of the method using appropriate drifting problems. In particular, we analyze the regularizing effect of changing the number of classifiers in the sequence or adapting the strength of the coupling. A comparison with other methods in several problems, including the well-known STAGGER dataset and the real-world electricity pricing domain, shows the good performance of TA-SVM in all tested situations.
publishDate 2011
dc.date.none.fl_str_mv 2011-01
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_6501
info:ar-repo/semantics/articulo
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/11336/15248
Grinblat, Guillermo Luis; Uzal, Lucas César; Ceccatto, Hermenegildo A.; Granitto, Pablo Miguel; Solving nonstationary classification problems with coupled support vector machines; Institute Of Electrical And Electronics Engineers; Ieee Transactions On Neural Networks; 22; 1; 1-2011; 37-51
1045-9227
1941-0093
url http://hdl.handle.net/11336/15248
identifier_str_mv Grinblat, Guillermo Luis; Uzal, Lucas César; Ceccatto, Hermenegildo A.; Granitto, Pablo Miguel; Solving nonstationary classification problems with coupled support vector machines; Institute Of Electrical And Electronics Engineers; Ieee Transactions On Neural Networks; 22; 1; 1-2011; 37-51
1045-9227
1941-0093
dc.language.none.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv info:eu-repo/semantics/altIdentifier/doi/10.1109/TNN.2010.2083684
info:eu-repo/semantics/altIdentifier/url/http://ieeexplore.ieee.org/document/5624639/?tp=&arnumber=5624639
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
eu_rights_str_mv openAccess
rights_invalid_str_mv https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
dc.format.none.fl_str_mv application/pdf
application/pdf
application/pdf
application/pdf
dc.publisher.none.fl_str_mv Institute Of Electrical And Electronics Engineers
publisher.none.fl_str_mv Institute Of Electrical And Electronics Engineers
dc.source.none.fl_str_mv reponame:CONICET Digital (CONICET)
instname:Consejo Nacional de Investigaciones Científicas y Técnicas
reponame_str CONICET Digital (CONICET)
collection CONICET Digital (CONICET)
instname_str Consejo Nacional de Investigaciones Científicas y Técnicas
repository.name.fl_str_mv CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicas
repository.mail.fl_str_mv dasensio@conicet.gov.ar; lcarlino@conicet.gov.ar
_version_ 1846083093972123648
score 13.22299