Active Semi-Supervised Learning using Particle Competition and Cooperation in Networks

Detalhes bibliográficos
Autor(a) principal: Breve, Fabricio
Data de Publicação: 2013
Outros Autores: IEEE
Tipo de documento: Artigo de conferência
Idioma: eng
Título da fonte: Repositório Institucional da UNESP
Texto Completo: http://hdl.handle.net/11449/196070
Resumo: Both Active Learning and Semi-Supervised Learning are important techniques when labeled data are scarce and unlabeled data are abundant. In this paper, these two machine learning techniques are combined into a new natureinspired method, which employs particles walking in networks generated from the data. It uses combined competitive and cooperative behavior in order to possess nodes of the network, and thus labeling the corresponding data items. Particles represent labeled nodes, and new particles can be added on the fly to the network as the result of queries (new labels). This built-in mechanism saves a lot of execution time comparing to active learning frameworks, since only nodes affected by the new particles are updated, i.e., the algorithm does not have to be executed again for each new query (or new set of queries). The algorithm naturally adapts itself to new scenarios, i.e., more particles and more labeled nodes. Experimental results on some real-world data sets are presented and the proposed active semisupervised learning method shows better classification accuracy than its only semi-supervised learning counterpart when the same amount of labeled data is used. Some criteria for selecting the rule to be used to choose data items to be queried are also identified.
id UNSP_225c931554c9f566ba3b40b38f2135c8
oai_identifier_str oai:repositorio.unesp.br:11449/196070
network_acronym_str UNSP
network_name_str Repositório Institucional da UNESP
repository_id_str 2946
spelling Active Semi-Supervised Learning using Particle Competition and Cooperation in NetworksBoth Active Learning and Semi-Supervised Learning are important techniques when labeled data are scarce and unlabeled data are abundant. In this paper, these two machine learning techniques are combined into a new natureinspired method, which employs particles walking in networks generated from the data. It uses combined competitive and cooperative behavior in order to possess nodes of the network, and thus labeling the corresponding data items. Particles represent labeled nodes, and new particles can be added on the fly to the network as the result of queries (new labels). This built-in mechanism saves a lot of execution time comparing to active learning frameworks, since only nodes affected by the new particles are updated, i.e., the algorithm does not have to be executed again for each new query (or new set of queries). The algorithm naturally adapts itself to new scenarios, i.e., more particles and more labeled nodes. Experimental results on some real-world data sets are presented and the proposed active semisupervised learning method shows better classification accuracy than its only semi-supervised learning counterpart when the same amount of labeled data is used. Some criteria for selecting the rule to be used to choose data items to be queried are also identified.Sao Paulo State Univ UNESP, Dept Stat Appl Math & Computat DEMAC, Inst Geosci & Exact Sci IGCE, Sao Paulo, BrazilSao Paulo State Univ UNESP, Dept Stat Appl Math & Computat DEMAC, Inst Geosci & Exact Sci IGCE, Sao Paulo, BrazilIeeeUniversidade Estadual Paulista (Unesp)Breve, FabricioIEEE2020-12-10T19:32:20Z2020-12-10T19:32:20Z2013-01-01info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/conferenceObject62013 International Joint Conference On Neural Networks (ijcnn). New York: Ieee, 6 p., 2013.2161-4393http://hdl.handle.net/11449/196070WOS:000349557200242Web of Sciencereponame:Repositório Institucional da UNESPinstname:Universidade Estadual Paulista (UNESP)instacron:UNESPeng2013 International Joint Conference On Neural Networks (ijcnn)info:eu-repo/semantics/openAccess2021-10-23T03:12:25Zoai:repositorio.unesp.br:11449/196070Repositório InstitucionalPUBhttp://repositorio.unesp.br/oai/requestopendoar:29462024-08-05T14:29:07.085599Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)false
dc.title.none.fl_str_mv Active Semi-Supervised Learning using Particle Competition and Cooperation in Networks
title Active Semi-Supervised Learning using Particle Competition and Cooperation in Networks
spellingShingle Active Semi-Supervised Learning using Particle Competition and Cooperation in Networks
Breve, Fabricio
title_short Active Semi-Supervised Learning using Particle Competition and Cooperation in Networks
title_full Active Semi-Supervised Learning using Particle Competition and Cooperation in Networks
title_fullStr Active Semi-Supervised Learning using Particle Competition and Cooperation in Networks
title_full_unstemmed Active Semi-Supervised Learning using Particle Competition and Cooperation in Networks
title_sort Active Semi-Supervised Learning using Particle Competition and Cooperation in Networks
author Breve, Fabricio
author_facet Breve, Fabricio
IEEE
author_role author
author2 IEEE
author2_role author
dc.contributor.none.fl_str_mv Universidade Estadual Paulista (Unesp)
dc.contributor.author.fl_str_mv Breve, Fabricio
IEEE
description Both Active Learning and Semi-Supervised Learning are important techniques when labeled data are scarce and unlabeled data are abundant. In this paper, these two machine learning techniques are combined into a new natureinspired method, which employs particles walking in networks generated from the data. It uses combined competitive and cooperative behavior in order to possess nodes of the network, and thus labeling the corresponding data items. Particles represent labeled nodes, and new particles can be added on the fly to the network as the result of queries (new labels). This built-in mechanism saves a lot of execution time comparing to active learning frameworks, since only nodes affected by the new particles are updated, i.e., the algorithm does not have to be executed again for each new query (or new set of queries). The algorithm naturally adapts itself to new scenarios, i.e., more particles and more labeled nodes. Experimental results on some real-world data sets are presented and the proposed active semisupervised learning method shows better classification accuracy than its only semi-supervised learning counterpart when the same amount of labeled data is used. Some criteria for selecting the rule to be used to choose data items to be queried are also identified.
publishDate 2013
dc.date.none.fl_str_mv 2013-01-01
2020-12-10T19:32:20Z
2020-12-10T19:32:20Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/conferenceObject
format conferenceObject
status_str publishedVersion
dc.identifier.uri.fl_str_mv 2013 International Joint Conference On Neural Networks (ijcnn). New York: Ieee, 6 p., 2013.
2161-4393
http://hdl.handle.net/11449/196070
WOS:000349557200242
identifier_str_mv 2013 International Joint Conference On Neural Networks (ijcnn). New York: Ieee, 6 p., 2013.
2161-4393
WOS:000349557200242
url http://hdl.handle.net/11449/196070
dc.language.iso.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv 2013 International Joint Conference On Neural Networks (ijcnn)
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv 6
dc.publisher.none.fl_str_mv Ieee
publisher.none.fl_str_mv Ieee
dc.source.none.fl_str_mv Web of Science
reponame:Repositório Institucional da UNESP
instname:Universidade Estadual Paulista (UNESP)
instacron:UNESP
instname_str Universidade Estadual Paulista (UNESP)
instacron_str UNESP
institution UNESP
reponame_str Repositório Institucional da UNESP
collection Repositório Institucional da UNESP
repository.name.fl_str_mv Repositório Institucional da UNESP - Universidade Estadual Paulista (UNESP)
repository.mail.fl_str_mv
_version_ 1808128365885915136