Using tensor products to detect unconditional label dependence in multilabel classifications

Multilabel (ML) classification tasks consist of assigning a set of labels to each input. It is well known that detecting label dependencies is crucial in order to improve the performance in ML problems. In this paper, we study a new kernel approach to take into account unconditional label dependence...

Descripción completa

Detalles Bibliográficos
Autores: Díez Peláez, Jorge|||0000-0002-1314-2441, Coz Velasco, Juan José del|||0000-0002-4288-3839, Luaces Rodríguez, Óscar|||0000-0001-8476-9412, Bahamonde Rionda, Antonio|||0000-0002-2188-9035
Tipo de recurso: artículo
Fecha de publicación:2016
País:España
Institución:Universidad de Oviedo (UNIOVI)
Repositorio:RUO. Repositorio Institucional de la Universidad de Oviedo
Idioma:inglés
OAI Identifier:oai:digibuo.uniovi.es:10651/35740
Acceso en línea:http://hdl.handle.net/10651/35740
https://dx.doi.org/10.1016/j.ins.2015.08.055
Access Level:acceso abierto
Palabra clave:Multilabel
Label dependence
Tensor products
Kernel methods
id ES_93cdabc0ec2cd5fa3a097ab9b9c55e89
oai_identifier_str oai:digibuo.uniovi.es:10651/35740
network_acronym_str ES
network_name_str España
repository_id_str
spelling Using tensor products to detect unconditional label dependence in multilabel classificationsDíez Peláez, Jorge|||0000-0002-1314-2441Coz Velasco, Juan José del|||0000-0002-4288-3839Luaces Rodríguez, Óscar|||0000-0001-8476-9412Bahamonde Rionda, Antonio|||0000-0002-2188-9035MultilabelLabel dependenceTensor productsKernel methodsMultilabel (ML) classification tasks consist of assigning a set of labels to each input. It is well known that detecting label dependencies is crucial in order to improve the performance in ML problems. In this paper, we study a new kernel approach to take into account unconditional label dependence between labels. The aim is to improve the performance measured by a micro-averaged loss function. The core idea is to transform a ML task into a binary classification problem whose inputs are drawn from a tensor space of the original input space and a representation of the labels. In this joint feature space we define a kernel to explicitly involve both labels and object descriptions. In addition to the theoretical contributions, the experimental results of this study provide an interesting conclusion: the performance in terms of Hamming Loss can be improved when unconditional label dependence is considered, as our method does. We report a thoroughly experimentation carried out with real world domains and several synthetic datasets devised to analyze the effect of exploiting label dependence in scenarios with different degrees of dependencyThe research reported here is supported in part under grant TIN2011-23558 from the MICINN (Ministerio de Economía y Competitividad, Spain)Elsevier20162016-02-01journal articlehttp://purl.org/coar/resource_type/c_6501AMhttp://purl.org/coar/version/c_ab4af688f83e57aainfo:eu-repo/semantics/articlehttp://hdl.handle.net/10651/35740https://dx.doi.org/10.1016/j.ins.2015.08.055reponame:RUO. Repositorio Institucional de la Universidad de Oviedoinstname:Universidad de Oviedo (UNIOVI)Inglésengopen accesshttp://purl.org/coar/access_right/c_abf2info:eu-repo/semantics/openAccessoai:digibuo.uniovi.es:10651/357402026-06-07T06:38:51Z
dc.title.none.fl_str_mv Using tensor products to detect unconditional label dependence in multilabel classifications
title Using tensor products to detect unconditional label dependence in multilabel classifications
spellingShingle Using tensor products to detect unconditional label dependence in multilabel classifications
Díez Peláez, Jorge|||0000-0002-1314-2441
Multilabel
Label dependence
Tensor products
Kernel methods
title_short Using tensor products to detect unconditional label dependence in multilabel classifications
title_full Using tensor products to detect unconditional label dependence in multilabel classifications
title_fullStr Using tensor products to detect unconditional label dependence in multilabel classifications
title_full_unstemmed Using tensor products to detect unconditional label dependence in multilabel classifications
title_sort Using tensor products to detect unconditional label dependence in multilabel classifications
dc.creator.none.fl_str_mv Díez Peláez, Jorge|||0000-0002-1314-2441
Coz Velasco, Juan José del|||0000-0002-4288-3839
Luaces Rodríguez, Óscar|||0000-0001-8476-9412
Bahamonde Rionda, Antonio|||0000-0002-2188-9035
author Díez Peláez, Jorge|||0000-0002-1314-2441
author_facet Díez Peláez, Jorge|||0000-0002-1314-2441
Coz Velasco, Juan José del|||0000-0002-4288-3839
Luaces Rodríguez, Óscar|||0000-0001-8476-9412
Bahamonde Rionda, Antonio|||0000-0002-2188-9035
author_role author
author2 Coz Velasco, Juan José del|||0000-0002-4288-3839
Luaces Rodríguez, Óscar|||0000-0001-8476-9412
Bahamonde Rionda, Antonio|||0000-0002-2188-9035
author2_role author
author
author
dc.subject.none.fl_str_mv Multilabel
Label dependence
Tensor products
Kernel methods
topic Multilabel
Label dependence
Tensor products
Kernel methods
description Multilabel (ML) classification tasks consist of assigning a set of labels to each input. It is well known that detecting label dependencies is crucial in order to improve the performance in ML problems. In this paper, we study a new kernel approach to take into account unconditional label dependence between labels. The aim is to improve the performance measured by a micro-averaged loss function. The core idea is to transform a ML task into a binary classification problem whose inputs are drawn from a tensor space of the original input space and a representation of the labels. In this joint feature space we define a kernel to explicitly involve both labels and object descriptions. In addition to the theoretical contributions, the experimental results of this study provide an interesting conclusion: the performance in terms of Hamming Loss can be improved when unconditional label dependence is considered, as our method does. We report a thoroughly experimentation carried out with real world domains and several synthetic datasets devised to analyze the effect of exploiting label dependence in scenarios with different degrees of dependency
publishDate 2016
dc.date.none.fl_str_mv 2016
2016-02-01
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
AM
http://purl.org/coar/version/c_ab4af688f83e57aa
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv http://hdl.handle.net/10651/35740
https://dx.doi.org/10.1016/j.ins.2015.08.055
url http://hdl.handle.net/10651/35740
https://dx.doi.org/10.1016/j.ins.2015.08.055
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
eu_rights_str_mv openAccess
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:RUO. Repositorio Institucional de la Universidad de Oviedo
instname:Universidad de Oviedo (UNIOVI)
instname_str Universidad de Oviedo (UNIOVI)
reponame_str RUO. Repositorio Institucional de la Universidad de Oviedo
collection RUO. Repositorio Institucional de la Universidad de Oviedo
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869413628109651968
score 15,301629

Similares en LA Referencia