Neural networks for feature-extraction in multi-target classification
Autor(a) principal: | |
---|---|
Data de Publicação: | 2020 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFSCAR |
Texto Completo: | https://repositorio.ufscar.br/handle/ufscar/13795 |
Resumo: | Multi-target learning is a prediction task where each data example is associated with multiple target-variables (outputs) simultaneously. One of the challenges in this research field is related to the high dimensionality of data present in multi-target datasets, and also the high number of target variables having dependencies among themselves. In such scenarios, it is crucial to extract lower-dimensional representations from the original input-space, such that these can be provided as input to other multi-target predictors. In this research, we proposed the use of Auto-Encoders and Restricted Boltzmann Machines as feature extractors in several multi-target classification datasets publicly available. Results were evaluated considering state-of-the-art multi-target classification methods and evaluation measures in the literature. The experiments showed that the neural networks were able to keep the predictive performance even when the extracted features corresponded to a dimension size equivalent to 10% of the original number of features and, in some cases, getting better results than the original datasets. |
id |
SCAR_201a7e7c97a181828b198cf1adc50439 |
---|---|
oai_identifier_str |
oai:repositorio.ufscar.br:ufscar/13795 |
network_acronym_str |
SCAR |
network_name_str |
Repositório Institucional da UFSCAR |
repository_id_str |
4322 |
spelling |
Cambuí, Brendon GouveiaCerri, Ricardohttp://lattes.cnpq.br/6266519868438512http://lattes.cnpq.br/0863602515011239c1768bc9-3305-4ce0-a190-e1442e91b7f12021-02-01T11:49:08Z2021-02-01T11:49:08Z2020-08-21CAMBUÍ, Brendon Gouveia. Neural networks for feature-extraction in multi-target classification. 2020. Dissertação (Mestrado em Ciência da Computação) – Universidade Federal de São Carlos, São Carlos, 2020. Disponível em: https://repositorio.ufscar.br/handle/ufscar/13795.https://repositorio.ufscar.br/handle/ufscar/13795Multi-target learning is a prediction task where each data example is associated with multiple target-variables (outputs) simultaneously. One of the challenges in this research field is related to the high dimensionality of data present in multi-target datasets, and also the high number of target variables having dependencies among themselves. In such scenarios, it is crucial to extract lower-dimensional representations from the original input-space, such that these can be provided as input to other multi-target predictors. In this research, we proposed the use of Auto-Encoders and Restricted Boltzmann Machines as feature extractors in several multi-target classification datasets publicly available. Results were evaluated considering state-of-the-art multi-target classification methods and evaluation measures in the literature. The experiments showed that the neural networks were able to keep the predictive performance even when the extracted features corresponded to a dimension size equivalent to 10% of the original number of features and, in some cases, getting better results than the original datasets.O aprendizado Multi-Target é uma tarefa de predição onde cada exemplo de um conjunto de dados é associado à múltiplas variáveis de saída simultaneamente. Um dos desafios desta pesquisa está associado à alta dimensionalidade dos dados presentes nos conjuntos Multi-Target, e o alto número de variáveis de saída que possuem dependência entre si. Nestes cenários, é crucial extrair representações de dimensões menores que a presente no conjunto de dados original, de forma que essas representações possam serem utilizadas como dados de entrada para os preditores Multi-Target. Nesta pesquisa, propomos o uso de Auto-Encoders e Restricted Boltzmann Machine como extratores de features em diversos datasets de classificação Multi-Target disponíveis ao domínio público. Os resultados foram avaliados considerando os métodos de classificação Multi-Target de estado-da-arte e os métodos de avaliação disponíveis na literatura. Os experimentos mostraram que as redes neurais foram capazes de manter uma performance preditiva competitiva, mesmo quando as features extraídas correspondiam a uma dimensão equivalente à 10% do número de features original e, em alguns casos, obtendo melhores resultados do que os obtidos utilizando os datasets originais.Não recebi financiamentoengUniversidade Federal de São CarlosCâmpus São CarlosPrograma de Pós-Graduação em Ciência da Computação - PPGCCUFSCarAttribution-NonCommercial-NoDerivs 3.0 Brazilhttp://creativecommons.org/licenses/by-nc-nd/3.0/br/info:eu-repo/semantics/openAccessMulti-Target ClassificationAuto-encodersRestricted Boltzmann MachineFeature-extractionDimensionality reductionCIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAONeural networks for feature-extraction in multi-target classificationRedes neurais para extração de features em classificação multi-targetinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesis600600c997f5ee-db84-40ed-8971-521dd105f2d1reponame:Repositório Institucional da UFSCARinstname:Universidade Federal de São Carlos (UFSCAR)instacron:UFSCARORIGINALDisserta__o_Mestrado___Brendon__Corre__o_.pdfDisserta__o_Mestrado___Brendon__Corre__o_.pdfDissertaçãoapplication/pdf4707669https://repositorio.ufscar.br/bitstream/ufscar/13795/1/Disserta__o_Mestrado___Brendon__Corre__o_.pdf596431b46c6763386e63ed331edcee89MD51PPGCC_Template_dec_BCO.pdfPPGCC_Template_dec_BCO.pdfAutorização do orientadorapplication/pdf124489https://repositorio.ufscar.br/bitstream/ufscar/13795/2/PPGCC_Template_dec_BCO.pdfaf443d9e6a2f89fbdd21294e6124db89MD52CC-LICENSElicense_rdflicense_rdfapplication/rdf+xml; charset=utf-8811https://repositorio.ufscar.br/bitstream/ufscar/13795/3/license_rdfe39d27027a6cc9cb039ad269a5db8e34MD53TEXTDisserta__o_Mestrado___Brendon__Corre__o_.pdf.txtDisserta__o_Mestrado___Brendon__Corre__o_.pdf.txtExtracted texttext/plain284905https://repositorio.ufscar.br/bitstream/ufscar/13795/4/Disserta__o_Mestrado___Brendon__Corre__o_.pdf.txt4ea0842a0ddb1c6137450d5230e13ef4MD54PPGCC_Template_dec_BCO.pdf.txtPPGCC_Template_dec_BCO.pdf.txtExtracted texttext/plain1594https://repositorio.ufscar.br/bitstream/ufscar/13795/6/PPGCC_Template_dec_BCO.pdf.txte571ca266a4ac8cec6207ca562f126ceMD56THUMBNAILDisserta__o_Mestrado___Brendon__Corre__o_.pdf.jpgDisserta__o_Mestrado___Brendon__Corre__o_.pdf.jpgIM Thumbnailimage/jpeg4711https://repositorio.ufscar.br/bitstream/ufscar/13795/5/Disserta__o_Mestrado___Brendon__Corre__o_.pdf.jpgb85fbb7471bf7f11ee69c00726202ae5MD55PPGCC_Template_dec_BCO.pdf.jpgPPGCC_Template_dec_BCO.pdf.jpgIM Thumbnailimage/jpeg13406https://repositorio.ufscar.br/bitstream/ufscar/13795/7/PPGCC_Template_dec_BCO.pdf.jpg64688499628c3e043892bb653f79b85eMD57ufscar/137952023-09-18 18:32:06.262oai:repositorio.ufscar.br:ufscar/13795Repositório InstitucionalPUBhttps://repositorio.ufscar.br/oai/requestopendoar:43222023-09-18T18:32:06Repositório Institucional da UFSCAR - Universidade Federal de São Carlos (UFSCAR)false |
dc.title.eng.fl_str_mv |
Neural networks for feature-extraction in multi-target classification |
dc.title.alternative.por.fl_str_mv |
Redes neurais para extração de features em classificação multi-target |
title |
Neural networks for feature-extraction in multi-target classification |
spellingShingle |
Neural networks for feature-extraction in multi-target classification Cambuí, Brendon Gouveia Multi-Target Classification Auto-encoders Restricted Boltzmann Machine Feature-extraction Dimensionality reduction CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO |
title_short |
Neural networks for feature-extraction in multi-target classification |
title_full |
Neural networks for feature-extraction in multi-target classification |
title_fullStr |
Neural networks for feature-extraction in multi-target classification |
title_full_unstemmed |
Neural networks for feature-extraction in multi-target classification |
title_sort |
Neural networks for feature-extraction in multi-target classification |
author |
Cambuí, Brendon Gouveia |
author_facet |
Cambuí, Brendon Gouveia |
author_role |
author |
dc.contributor.authorlattes.por.fl_str_mv |
http://lattes.cnpq.br/0863602515011239 |
dc.contributor.author.fl_str_mv |
Cambuí, Brendon Gouveia |
dc.contributor.advisor1.fl_str_mv |
Cerri, Ricardo |
dc.contributor.advisor1Lattes.fl_str_mv |
http://lattes.cnpq.br/6266519868438512 |
dc.contributor.authorID.fl_str_mv |
c1768bc9-3305-4ce0-a190-e1442e91b7f1 |
contributor_str_mv |
Cerri, Ricardo |
dc.subject.eng.fl_str_mv |
Multi-Target Classification Auto-encoders Restricted Boltzmann Machine Feature-extraction Dimensionality reduction |
topic |
Multi-Target Classification Auto-encoders Restricted Boltzmann Machine Feature-extraction Dimensionality reduction CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO |
dc.subject.cnpq.fl_str_mv |
CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO |
description |
Multi-target learning is a prediction task where each data example is associated with multiple target-variables (outputs) simultaneously. One of the challenges in this research field is related to the high dimensionality of data present in multi-target datasets, and also the high number of target variables having dependencies among themselves. In such scenarios, it is crucial to extract lower-dimensional representations from the original input-space, such that these can be provided as input to other multi-target predictors. In this research, we proposed the use of Auto-Encoders and Restricted Boltzmann Machines as feature extractors in several multi-target classification datasets publicly available. Results were evaluated considering state-of-the-art multi-target classification methods and evaluation measures in the literature. The experiments showed that the neural networks were able to keep the predictive performance even when the extracted features corresponded to a dimension size equivalent to 10% of the original number of features and, in some cases, getting better results than the original datasets. |
publishDate |
2020 |
dc.date.issued.fl_str_mv |
2020-08-21 |
dc.date.accessioned.fl_str_mv |
2021-02-01T11:49:08Z |
dc.date.available.fl_str_mv |
2021-02-01T11:49:08Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.citation.fl_str_mv |
CAMBUÍ, Brendon Gouveia. Neural networks for feature-extraction in multi-target classification. 2020. Dissertação (Mestrado em Ciência da Computação) – Universidade Federal de São Carlos, São Carlos, 2020. Disponível em: https://repositorio.ufscar.br/handle/ufscar/13795. |
dc.identifier.uri.fl_str_mv |
https://repositorio.ufscar.br/handle/ufscar/13795 |
identifier_str_mv |
CAMBUÍ, Brendon Gouveia. Neural networks for feature-extraction in multi-target classification. 2020. Dissertação (Mestrado em Ciência da Computação) – Universidade Federal de São Carlos, São Carlos, 2020. Disponível em: https://repositorio.ufscar.br/handle/ufscar/13795. |
url |
https://repositorio.ufscar.br/handle/ufscar/13795 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.confidence.fl_str_mv |
600 600 |
dc.relation.authority.fl_str_mv |
c997f5ee-db84-40ed-8971-521dd105f2d1 |
dc.rights.driver.fl_str_mv |
Attribution-NonCommercial-NoDerivs 3.0 Brazil http://creativecommons.org/licenses/by-nc-nd/3.0/br/ info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
Attribution-NonCommercial-NoDerivs 3.0 Brazil http://creativecommons.org/licenses/by-nc-nd/3.0/br/ |
eu_rights_str_mv |
openAccess |
dc.publisher.none.fl_str_mv |
Universidade Federal de São Carlos Câmpus São Carlos |
dc.publisher.program.fl_str_mv |
Programa de Pós-Graduação em Ciência da Computação - PPGCC |
dc.publisher.initials.fl_str_mv |
UFSCar |
publisher.none.fl_str_mv |
Universidade Federal de São Carlos Câmpus São Carlos |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UFSCAR instname:Universidade Federal de São Carlos (UFSCAR) instacron:UFSCAR |
instname_str |
Universidade Federal de São Carlos (UFSCAR) |
instacron_str |
UFSCAR |
institution |
UFSCAR |
reponame_str |
Repositório Institucional da UFSCAR |
collection |
Repositório Institucional da UFSCAR |
bitstream.url.fl_str_mv |
https://repositorio.ufscar.br/bitstream/ufscar/13795/1/Disserta__o_Mestrado___Brendon__Corre__o_.pdf https://repositorio.ufscar.br/bitstream/ufscar/13795/2/PPGCC_Template_dec_BCO.pdf https://repositorio.ufscar.br/bitstream/ufscar/13795/3/license_rdf https://repositorio.ufscar.br/bitstream/ufscar/13795/4/Disserta__o_Mestrado___Brendon__Corre__o_.pdf.txt https://repositorio.ufscar.br/bitstream/ufscar/13795/6/PPGCC_Template_dec_BCO.pdf.txt https://repositorio.ufscar.br/bitstream/ufscar/13795/5/Disserta__o_Mestrado___Brendon__Corre__o_.pdf.jpg https://repositorio.ufscar.br/bitstream/ufscar/13795/7/PPGCC_Template_dec_BCO.pdf.jpg |
bitstream.checksum.fl_str_mv |
596431b46c6763386e63ed331edcee89 af443d9e6a2f89fbdd21294e6124db89 e39d27027a6cc9cb039ad269a5db8e34 4ea0842a0ddb1c6137450d5230e13ef4 e571ca266a4ac8cec6207ca562f126ce b85fbb7471bf7f11ee69c00726202ae5 64688499628c3e043892bb653f79b85e |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 MD5 MD5 MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional da UFSCAR - Universidade Federal de São Carlos (UFSCAR) |
repository.mail.fl_str_mv |
|
_version_ |
1802136378374356992 |