A novel hybrid feature selection algorithm for hierarchical classification
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | , , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFLA |
Texto Completo: | http://repositorio.ufla.br/jspui/handle/1/50685 |
Resumo: | Feature selection is a widespread preprocessing step in the data mining field. One of its purposes is to reduce the number of original dataset features to improve a predictive model’s performance. Despite the benefits of feature selection for the classification task, to the best of our knowledge, few studies in the literature address feature selection for the hierarchical classification context. This paper proposes a novel feature selection method based on the general variable neighborhood search metaheuristic, combining a filter and a wrapper step, wherein a global model hierarchical classifier evaluates feature subsets. We used twelve datasets from the proteins and images domains to perform computational experiments to validate the effect of the proposed algorithm on classification performance when using two global hierarchical classifiers proposed in the literature. Statistical tests showed that using our method for feature selection led to predictive performances that were consistently better than or equivalent to that obtained by using all features with the benefit of reducing the number of features needed, which justifies its efficiency for the hierarchical classification scenario. |
id |
UFLA_404742d28e69e66b83c8ae2a9ab5e97a |
---|---|
oai_identifier_str |
oai:localhost:1/50685 |
network_acronym_str |
UFLA |
network_name_str |
Repositório Institucional da UFLA |
repository_id_str |
|
spelling |
A novel hybrid feature selection algorithm for hierarchical classificationFeature selectionHierarchical single-label classificationVariable neighborhood searchWrapperSeleção de recursosClassificação hierárquica de rótulo únicoPesquisa variável de vizinhançaFeature selection is a widespread preprocessing step in the data mining field. One of its purposes is to reduce the number of original dataset features to improve a predictive model’s performance. Despite the benefits of feature selection for the classification task, to the best of our knowledge, few studies in the literature address feature selection for the hierarchical classification context. This paper proposes a novel feature selection method based on the general variable neighborhood search metaheuristic, combining a filter and a wrapper step, wherein a global model hierarchical classifier evaluates feature subsets. We used twelve datasets from the proteins and images domains to perform computational experiments to validate the effect of the proposed algorithm on classification performance when using two global hierarchical classifiers proposed in the literature. Statistical tests showed that using our method for feature selection led to predictive performances that were consistently better than or equivalent to that obtained by using all features with the benefit of reducing the number of features needed, which justifies its efficiency for the hierarchical classification scenario.IEEE Xplore2022-07-21T21:55:50Z2022-07-21T21:55:50Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfLIMA, H. C. S. C. et al. A novel hybrid feature selection algorithm for hierarchical classification. IEEE Access, [S. l.], v. 9, p. 127278-127292, 202. DOI: 10.1109/ACCESS.2021.3112396.http://repositorio.ufla.br/jspui/handle/1/50685IEEE Accessreponame:Repositório Institucional da UFLAinstname:Universidade Federal de Lavras (UFLA)instacron:UFLAAttribution 4.0 Internationalhttp://creativecommons.org/licenses/by/4.0/info:eu-repo/semantics/openAccessLima, Helen C. S. C.Otero, Fernando E. B.Merschmann, Luiz H. C.Souza, Marcone J. F.eng2023-05-03T13:09:24Zoai:localhost:1/50685Repositório InstitucionalPUBhttp://repositorio.ufla.br/oai/requestnivaldo@ufla.br || repositorio.biblioteca@ufla.bropendoar:2023-05-03T13:09:24Repositório Institucional da UFLA - Universidade Federal de Lavras (UFLA)false |
dc.title.none.fl_str_mv |
A novel hybrid feature selection algorithm for hierarchical classification |
title |
A novel hybrid feature selection algorithm for hierarchical classification |
spellingShingle |
A novel hybrid feature selection algorithm for hierarchical classification Lima, Helen C. S. C. Feature selection Hierarchical single-label classification Variable neighborhood search Wrapper Seleção de recursos Classificação hierárquica de rótulo único Pesquisa variável de vizinhança |
title_short |
A novel hybrid feature selection algorithm for hierarchical classification |
title_full |
A novel hybrid feature selection algorithm for hierarchical classification |
title_fullStr |
A novel hybrid feature selection algorithm for hierarchical classification |
title_full_unstemmed |
A novel hybrid feature selection algorithm for hierarchical classification |
title_sort |
A novel hybrid feature selection algorithm for hierarchical classification |
author |
Lima, Helen C. S. C. |
author_facet |
Lima, Helen C. S. C. Otero, Fernando E. B. Merschmann, Luiz H. C. Souza, Marcone J. F. |
author_role |
author |
author2 |
Otero, Fernando E. B. Merschmann, Luiz H. C. Souza, Marcone J. F. |
author2_role |
author author author |
dc.contributor.author.fl_str_mv |
Lima, Helen C. S. C. Otero, Fernando E. B. Merschmann, Luiz H. C. Souza, Marcone J. F. |
dc.subject.por.fl_str_mv |
Feature selection Hierarchical single-label classification Variable neighborhood search Wrapper Seleção de recursos Classificação hierárquica de rótulo único Pesquisa variável de vizinhança |
topic |
Feature selection Hierarchical single-label classification Variable neighborhood search Wrapper Seleção de recursos Classificação hierárquica de rótulo único Pesquisa variável de vizinhança |
description |
Feature selection is a widespread preprocessing step in the data mining field. One of its purposes is to reduce the number of original dataset features to improve a predictive model’s performance. Despite the benefits of feature selection for the classification task, to the best of our knowledge, few studies in the literature address feature selection for the hierarchical classification context. This paper proposes a novel feature selection method based on the general variable neighborhood search metaheuristic, combining a filter and a wrapper step, wherein a global model hierarchical classifier evaluates feature subsets. We used twelve datasets from the proteins and images domains to perform computational experiments to validate the effect of the proposed algorithm on classification performance when using two global hierarchical classifiers proposed in the literature. Statistical tests showed that using our method for feature selection led to predictive performances that were consistently better than or equivalent to that obtained by using all features with the benefit of reducing the number of features needed, which justifies its efficiency for the hierarchical classification scenario. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-07-21T21:55:50Z 2022-07-21T21:55:50Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
LIMA, H. C. S. C. et al. A novel hybrid feature selection algorithm for hierarchical classification. IEEE Access, [S. l.], v. 9, p. 127278-127292, 202. DOI: 10.1109/ACCESS.2021.3112396. http://repositorio.ufla.br/jspui/handle/1/50685 |
identifier_str_mv |
LIMA, H. C. S. C. et al. A novel hybrid feature selection algorithm for hierarchical classification. IEEE Access, [S. l.], v. 9, p. 127278-127292, 202. DOI: 10.1109/ACCESS.2021.3112396. |
url |
http://repositorio.ufla.br/jspui/handle/1/50685 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
Attribution 4.0 International http://creativecommons.org/licenses/by/4.0/ info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Attribution 4.0 International http://creativecommons.org/licenses/by/4.0/ |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
IEEE Xplore |
publisher.none.fl_str_mv |
IEEE Xplore |
dc.source.none.fl_str_mv |
IEEE Access reponame:Repositório Institucional da UFLA instname:Universidade Federal de Lavras (UFLA) instacron:UFLA |
instname_str |
Universidade Federal de Lavras (UFLA) |
instacron_str |
UFLA |
institution |
UFLA |
reponame_str |
Repositório Institucional da UFLA |
collection |
Repositório Institucional da UFLA |
repository.name.fl_str_mv |
Repositório Institucional da UFLA - Universidade Federal de Lavras (UFLA) |
repository.mail.fl_str_mv |
nivaldo@ufla.br || repositorio.biblioteca@ufla.br |
_version_ |
1807835126626779136 |