Variants of the Fast Adaptive Stacking of Ensembles algorithm
Autor(a) principal: | |
---|---|
Data de Publicação: | 2019 |
Tipo de documento: | Dissertação |
Idioma: | por |
Título da fonte: | Repositório Institucional da UFPE |
Texto Completo: | https://repositorio.ufpe.br/handle/123456789/36042 |
Resumo: | The treatment of large data streams in the presence of concept drifts is one of the main challenges in the fields of machine learning and data mining. This dissertation presents two families of ensemble algorithms designed to quickly adapt to concept drifts, both abrupt and gradual. The families Fast Stacking of Ensembles boosting the Old (FASEO) and Fast Stacking of Ensembles boosting the Best (FASEB) are adaptations of the Fast Adaptive Stacking of Ensembles (FASE) algorithm, designed to improve run-time and memory requirements, without presenting a significant decrease in terms of accuracy when compared to the original FASE. In order to achieve a more efficient model, adjustments were made in the update strategy and voting procedure of the ensemble. To evaluate the proposals against state of the art methods, Naïve Bayes (NB) and Hoeffding Tree (HT) are used, as learners, to compare the performance of the algorithms on artificial and realworld data-sets. An extensive experimental investigation with a total of 70 experiments and application of Friedman and Nemenyi statistical tests showed the families FASEO and FASEB are more efficient than FASE with respect to execution time and memory in many scenarios, often also achieving better accuracy results. |
id |
UFPE_f52538aaaaa4291adf0d7d727caaa03f |
---|---|
oai_identifier_str |
oai:repositorio.ufpe.br:123456789/36042 |
network_acronym_str |
UFPE |
network_name_str |
Repositório Institucional da UFPE |
repository_id_str |
2221 |
spelling |
MARIÑO, Laura María Palominohttp://lattes.cnpq.br/4050952327940886http://lattes.cnpq.br/5943634209341438VASCONCELOS, Germano Crispim2020-01-17T12:03:06Z2020-01-17T12:03:06Z2019-07-26MARIÑO, Laura María Palomino Variants of the Fast Adaptive Stacking of Ensembles algorithm. 2019. Dissertação (Mestrado em Ciência da Computação) – Universidade Federal de Pernambuco, Recife, 2019.https://repositorio.ufpe.br/handle/123456789/36042The treatment of large data streams in the presence of concept drifts is one of the main challenges in the fields of machine learning and data mining. This dissertation presents two families of ensemble algorithms designed to quickly adapt to concept drifts, both abrupt and gradual. The families Fast Stacking of Ensembles boosting the Old (FASEO) and Fast Stacking of Ensembles boosting the Best (FASEB) are adaptations of the Fast Adaptive Stacking of Ensembles (FASE) algorithm, designed to improve run-time and memory requirements, without presenting a significant decrease in terms of accuracy when compared to the original FASE. In order to achieve a more efficient model, adjustments were made in the update strategy and voting procedure of the ensemble. To evaluate the proposals against state of the art methods, Naïve Bayes (NB) and Hoeffding Tree (HT) are used, as learners, to compare the performance of the algorithms on artificial and realworld data-sets. An extensive experimental investigation with a total of 70 experiments and application of Friedman and Nemenyi statistical tests showed the families FASEO and FASEB are more efficient than FASE with respect to execution time and memory in many scenarios, often also achieving better accuracy results.CAPESO tratamento de grandes fluxos de dados na presença de mudanças de conceito é um dos principais desafios nas áreas de aprendizado de máquina e mineração de dados. Essa dissertação apresenta duas famílias de algoritmos de combinação de classificadores projetados para se adaptar rapidamente a mudanças de conceitos, tanto abruptos quanto graduais. As famílias Fast Stacking of Ensembles boosting the Old (FASEO) e Fast Stacking of Ensembles boosting the Best (FASEB) são adaptações do algoritmo Fast Stacking of Ensembles (FASE), projetadas para melhorar seu tempo de execução e consumo de memória, sem apresentar uma diminuição significativa de desempenho em termos de acurácia em comparação com o algoritmo original. Para obter um modelo mais eficiente, foram feitos ajustes na estratégia de atualização e no processo de votação de FASE. Para avaliar as propostas em relação ao estado da arte, usamos o Naive Bayes (NB) e o Hoeffding Tree (HT) como classificadores base para comparar o desempenho dos algoritmos em conjuntos de dados reais e sintéticos. Um avaliação experimental extensa, com um total 70 experimentos e emprego dos testes estatísticos de Friedman e Nemenyi, mostraram que as famílias FASEO e FASEB são mais eficientes que FASE com respeito a tempo de execução e memória em vários cenários, as vezes alcançando também melhores resultados na acurácia dos algoritmos.porUniversidade Federal de PernambucoPrograma de Pos Graduacao em Ciencia da ComputacaoUFPEBrasilAttribution-NonCommercial-NoDerivs 3.0 Brazilhttp://creativecommons.org/licenses/by-nc-nd/3.0/br/info:eu-repo/semantics/openAccessInteligência artificialMudanças de conceitoFluxo de dadosMétodos de Combinação de ClassificadoresVariants of the Fast Adaptive Stacking of Ensembles algorithminfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesismestradoreponame:Repositório Institucional da UFPEinstname:Universidade Federal de Pernambuco (UFPE)instacron:UFPEORIGINALDISSERTAÇÃO Laura María Palomino Mariño.pdfDISSERTAÇÃO Laura María Palomino Mariño.pdfapplication/pdf1816035https://repositorio.ufpe.br/bitstream/123456789/36042/1/DISSERTA%c3%87%c3%83O%20Laura%20Mar%c3%ada%20Palomino%20Mari%c3%b1o.pdf057b7fdd76b2f0f5c2e5699a64eb464fMD51CC-LICENSElicense_rdflicense_rdfapplication/rdf+xml; charset=utf-8811https://repositorio.ufpe.br/bitstream/123456789/36042/2/license_rdfe39d27027a6cc9cb039ad269a5db8e34MD52LICENSElicense.txtlicense.txttext/plain; charset=utf-82310https://repositorio.ufpe.br/bitstream/123456789/36042/3/license.txtbd573a5ca8288eb7272482765f819534MD53TEXTDISSERTAÇÃO Laura María Palomino Mariño.pdf.txtDISSERTAÇÃO Laura María Palomino Mariño.pdf.txtExtracted texttext/plain275286https://repositorio.ufpe.br/bitstream/123456789/36042/4/DISSERTA%c3%87%c3%83O%20Laura%20Mar%c3%ada%20Palomino%20Mari%c3%b1o.pdf.txt952b46cc1449bdf3452020e3a96f5c39MD54THUMBNAILDISSERTAÇÃO Laura María Palomino Mariño.pdf.jpgDISSERTAÇÃO Laura María Palomino Mariño.pdf.jpgGenerated Thumbnailimage/jpeg1238https://repositorio.ufpe.br/bitstream/123456789/36042/5/DISSERTA%c3%87%c3%83O%20Laura%20Mar%c3%ada%20Palomino%20Mari%c3%b1o.pdf.jpg56552aa8f57be54fdc08ba900e6aec61MD55123456789/360422020-01-18 02:15:58.832oai:repositorio.ufpe.br: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ório InstitucionalPUBhttps://repositorio.ufpe.br/oai/requestattena@ufpe.bropendoar:22212020-01-18T05:15:58Repositório Institucional da UFPE - Universidade Federal de Pernambuco (UFPE)false |
dc.title.pt_BR.fl_str_mv |
Variants of the Fast Adaptive Stacking of Ensembles algorithm |
title |
Variants of the Fast Adaptive Stacking of Ensembles algorithm |
spellingShingle |
Variants of the Fast Adaptive Stacking of Ensembles algorithm MARIÑO, Laura María Palomino Inteligência artificial Mudanças de conceito Fluxo de dados Métodos de Combinação de Classificadores |
title_short |
Variants of the Fast Adaptive Stacking of Ensembles algorithm |
title_full |
Variants of the Fast Adaptive Stacking of Ensembles algorithm |
title_fullStr |
Variants of the Fast Adaptive Stacking of Ensembles algorithm |
title_full_unstemmed |
Variants of the Fast Adaptive Stacking of Ensembles algorithm |
title_sort |
Variants of the Fast Adaptive Stacking of Ensembles algorithm |
author |
MARIÑO, Laura María Palomino |
author_facet |
MARIÑO, Laura María Palomino |
author_role |
author |
dc.contributor.authorLattes.pt_BR.fl_str_mv |
http://lattes.cnpq.br/4050952327940886 |
dc.contributor.advisorLattes.pt_BR.fl_str_mv |
http://lattes.cnpq.br/5943634209341438 |
dc.contributor.author.fl_str_mv |
MARIÑO, Laura María Palomino |
dc.contributor.advisor1.fl_str_mv |
VASCONCELOS, Germano Crispim |
contributor_str_mv |
VASCONCELOS, Germano Crispim |
dc.subject.por.fl_str_mv |
Inteligência artificial Mudanças de conceito Fluxo de dados Métodos de Combinação de Classificadores |
topic |
Inteligência artificial Mudanças de conceito Fluxo de dados Métodos de Combinação de Classificadores |
description |
The treatment of large data streams in the presence of concept drifts is one of the main challenges in the fields of machine learning and data mining. This dissertation presents two families of ensemble algorithms designed to quickly adapt to concept drifts, both abrupt and gradual. The families Fast Stacking of Ensembles boosting the Old (FASEO) and Fast Stacking of Ensembles boosting the Best (FASEB) are adaptations of the Fast Adaptive Stacking of Ensembles (FASE) algorithm, designed to improve run-time and memory requirements, without presenting a significant decrease in terms of accuracy when compared to the original FASE. In order to achieve a more efficient model, adjustments were made in the update strategy and voting procedure of the ensemble. To evaluate the proposals against state of the art methods, Naïve Bayes (NB) and Hoeffding Tree (HT) are used, as learners, to compare the performance of the algorithms on artificial and realworld data-sets. An extensive experimental investigation with a total of 70 experiments and application of Friedman and Nemenyi statistical tests showed the families FASEO and FASEB are more efficient than FASE with respect to execution time and memory in many scenarios, often also achieving better accuracy results. |
publishDate |
2019 |
dc.date.issued.fl_str_mv |
2019-07-26 |
dc.date.accessioned.fl_str_mv |
2020-01-17T12:03:06Z |
dc.date.available.fl_str_mv |
2020-01-17T12:03:06Z |
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 |
MARIÑO, Laura María Palomino Variants of the Fast Adaptive Stacking of Ensembles algorithm. 2019. Dissertação (Mestrado em Ciência da Computação) – Universidade Federal de Pernambuco, Recife, 2019. |
dc.identifier.uri.fl_str_mv |
https://repositorio.ufpe.br/handle/123456789/36042 |
identifier_str_mv |
MARIÑO, Laura María Palomino Variants of the Fast Adaptive Stacking of Ensembles algorithm. 2019. Dissertação (Mestrado em Ciência da Computação) – Universidade Federal de Pernambuco, Recife, 2019. |
url |
https://repositorio.ufpe.br/handle/123456789/36042 |
dc.language.iso.fl_str_mv |
por |
language |
por |
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 Pernambuco |
dc.publisher.program.fl_str_mv |
Programa de Pos Graduacao em Ciencia da Computacao |
dc.publisher.initials.fl_str_mv |
UFPE |
dc.publisher.country.fl_str_mv |
Brasil |
publisher.none.fl_str_mv |
Universidade Federal de Pernambuco |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UFPE instname:Universidade Federal de Pernambuco (UFPE) instacron:UFPE |
instname_str |
Universidade Federal de Pernambuco (UFPE) |
instacron_str |
UFPE |
institution |
UFPE |
reponame_str |
Repositório Institucional da UFPE |
collection |
Repositório Institucional da UFPE |
bitstream.url.fl_str_mv |
https://repositorio.ufpe.br/bitstream/123456789/36042/1/DISSERTA%c3%87%c3%83O%20Laura%20Mar%c3%ada%20Palomino%20Mari%c3%b1o.pdf https://repositorio.ufpe.br/bitstream/123456789/36042/2/license_rdf https://repositorio.ufpe.br/bitstream/123456789/36042/3/license.txt https://repositorio.ufpe.br/bitstream/123456789/36042/4/DISSERTA%c3%87%c3%83O%20Laura%20Mar%c3%ada%20Palomino%20Mari%c3%b1o.pdf.txt https://repositorio.ufpe.br/bitstream/123456789/36042/5/DISSERTA%c3%87%c3%83O%20Laura%20Mar%c3%ada%20Palomino%20Mari%c3%b1o.pdf.jpg |
bitstream.checksum.fl_str_mv |
057b7fdd76b2f0f5c2e5699a64eb464f e39d27027a6cc9cb039ad269a5db8e34 bd573a5ca8288eb7272482765f819534 952b46cc1449bdf3452020e3a96f5c39 56552aa8f57be54fdc08ba900e6aec61 |
bitstream.checksumAlgorithm.fl_str_mv |
MD5 MD5 MD5 MD5 MD5 |
repository.name.fl_str_mv |
Repositório Institucional da UFPE - Universidade Federal de Pernambuco (UFPE) |
repository.mail.fl_str_mv |
attena@ufpe.br |
_version_ |
1802310632392884224 |