Performance of combined models in discrete binary classification
Autor(a) principal: | |
---|---|
Data de Publicação: | 2017 |
Outros Autores: | , |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | http://hdl.handle.net/10071/13072 |
Resumo: | Diverse Discrete Discriminant Analysis (DDA) models perform differently in different samples. This fact has encouraged research in combined models which seems particularly promising when the a priori classes are not well separated or when small or moderate sized samples are considered, which often occurs in practice. In this study, we evaluate the performance of a convex combination of two DDA models: the First-Order Independence Model (FOIM) and the Dependence Trees Model (DTM). We use simulated data sets with two classes and consider diverse data complexity factors which may influence performance of the combined model -the separation of classes, balance, and number of missing states, as well as sample size and also the number of parameters to be estimated in DDA. We resort to cross-validation to evaluate the precision of classification. The results obtained illustrate the advantage of the proposed combination when compared with FOIM and DTM: it yields the best results, especially when very small samples are considered. The experimental study also provides a ranking of the data complexity factors, according to their relative impact on classification performance, by means of a regression model. It leads to the conclusion that the separation of classes is the most influential factor in classification performance. The ratio between the number of degrees of freedom and sample size, along with the proportion of missing states in the minority class, also has significant impact on classification performance. An additional gain of this study, also deriving from the estimated regression model, is the ability to successfully predict the precision of classification in a real data set based on the data complexity factors. |
id |
RCAP_e56692c4df942a7915fba014c7237ac8 |
---|---|
oai_identifier_str |
oai:repositorio.iscte-iul.pt:10071/13072 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Performance of combined models in discrete binary classificationClassification performanceCombined models for classificationDiscrete discriminant analysisSeparabilityDiverse Discrete Discriminant Analysis (DDA) models perform differently in different samples. This fact has encouraged research in combined models which seems particularly promising when the a priori classes are not well separated or when small or moderate sized samples are considered, which often occurs in practice. In this study, we evaluate the performance of a convex combination of two DDA models: the First-Order Independence Model (FOIM) and the Dependence Trees Model (DTM). We use simulated data sets with two classes and consider diverse data complexity factors which may influence performance of the combined model -the separation of classes, balance, and number of missing states, as well as sample size and also the number of parameters to be estimated in DDA. We resort to cross-validation to evaluate the precision of classification. The results obtained illustrate the advantage of the proposed combination when compared with FOIM and DTM: it yields the best results, especially when very small samples are considered. The experimental study also provides a ranking of the data complexity factors, according to their relative impact on classification performance, by means of a regression model. It leads to the conclusion that the separation of classes is the most influential factor in classification performance. The ratio between the number of degrees of freedom and sample size, along with the proportion of missing states in the minority class, also has significant impact on classification performance. An additional gain of this study, also deriving from the estimated regression model, is the ability to successfully predict the precision of classification in a real data set based on the data complexity factors.Hogrefe and Huber Publisher2017-04-20T15:21:27Z2017-01-01T00:00:00Z20172019-03-21T17:47:10Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfhttp://hdl.handle.net/10071/13072eng1614-188110.1027/1614-2241/a000117Marques, A.Ferreira, A. S.Cardoso, M. G. M. S.info:eu-repo/semantics/embargoedAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-11-09T17:51:24Zoai:repositorio.iscte-iul.pt:10071/13072Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T22:25:27.552711Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Performance of combined models in discrete binary classification |
title |
Performance of combined models in discrete binary classification |
spellingShingle |
Performance of combined models in discrete binary classification Marques, A. Classification performance Combined models for classification Discrete discriminant analysis Separability |
title_short |
Performance of combined models in discrete binary classification |
title_full |
Performance of combined models in discrete binary classification |
title_fullStr |
Performance of combined models in discrete binary classification |
title_full_unstemmed |
Performance of combined models in discrete binary classification |
title_sort |
Performance of combined models in discrete binary classification |
author |
Marques, A. |
author_facet |
Marques, A. Ferreira, A. S. Cardoso, M. G. M. S. |
author_role |
author |
author2 |
Ferreira, A. S. Cardoso, M. G. M. S. |
author2_role |
author author |
dc.contributor.author.fl_str_mv |
Marques, A. Ferreira, A. S. Cardoso, M. G. M. S. |
dc.subject.por.fl_str_mv |
Classification performance Combined models for classification Discrete discriminant analysis Separability |
topic |
Classification performance Combined models for classification Discrete discriminant analysis Separability |
description |
Diverse Discrete Discriminant Analysis (DDA) models perform differently in different samples. This fact has encouraged research in combined models which seems particularly promising when the a priori classes are not well separated or when small or moderate sized samples are considered, which often occurs in practice. In this study, we evaluate the performance of a convex combination of two DDA models: the First-Order Independence Model (FOIM) and the Dependence Trees Model (DTM). We use simulated data sets with two classes and consider diverse data complexity factors which may influence performance of the combined model -the separation of classes, balance, and number of missing states, as well as sample size and also the number of parameters to be estimated in DDA. We resort to cross-validation to evaluate the precision of classification. The results obtained illustrate the advantage of the proposed combination when compared with FOIM and DTM: it yields the best results, especially when very small samples are considered. The experimental study also provides a ranking of the data complexity factors, according to their relative impact on classification performance, by means of a regression model. It leads to the conclusion that the separation of classes is the most influential factor in classification performance. The ratio between the number of degrees of freedom and sample size, along with the proportion of missing states in the minority class, also has significant impact on classification performance. An additional gain of this study, also deriving from the estimated regression model, is the ability to successfully predict the precision of classification in a real data set based on the data complexity factors. |
publishDate |
2017 |
dc.date.none.fl_str_mv |
2017-04-20T15:21:27Z 2017-01-01T00:00:00Z 2017 2019-03-21T17:47:10Z |
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 |
http://hdl.handle.net/10071/13072 |
url |
http://hdl.handle.net/10071/13072 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
1614-1881 10.1027/1614-2241/a000117 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/embargoedAccess |
eu_rights_str_mv |
embargoedAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Hogrefe and Huber Publisher |
publisher.none.fl_str_mv |
Hogrefe and Huber Publisher |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799134817492664320 |