Functional-bandwidth kernel for Support Vector Machine with Functional Data_An alternating optimization algorithm

Functional Data Analysis (FDA) is devoted to the study of data which are functions. Support Vector Ma- chine (SVM) is a benchmark tool for classification, in particular, of functional data. SVM is frequently used with a kernel (e.g.: Gaussian) which involves a scalar bandwidth parameter. In this pap...

Full description

Bibliographic Details
Authors: Blanquero Bravo, Rafael, Carrizosa Priego, Emilio José, Jiménez Cordero, María Asunción, Martín Barragán, Belén
Format: article
Status:Published version
Publication Date:2018
Country:España
Institution:Universidad de Sevilla (US)
Repository:idUS. Depósito de Investigación de la Universidad de Sevilla
OAI Identifier:oai:idus.us.es:11441/107638
Online Access:https://hdl.handle.net/11441/107638
https://doi.org/10.1016/j.ejor.2018.11.024
Access Level:Open access
Keyword:Data mining
Functional Data classification
Parameter tuning
SVM
Functional bandwidth
id ES_56d4b08cca784f265cfd71a8adeb8ffe
oai_identifier_str oai:idus.us.es:11441/107638
network_acronym_str ES
network_name_str España
repository_id_str
spelling Functional-bandwidth kernel for Support Vector Machine with Functional Data_An alternating optimization algorithmBlanquero Bravo, RafaelCarrizosa Priego, Emilio JoséJiménez Cordero, María AsunciónMartín Barragán, BelénData miningFunctional Data classificationParameter tuningSVMFunctional bandwidthFunctional Data Analysis (FDA) is devoted to the study of data which are functions. Support Vector Ma- chine (SVM) is a benchmark tool for classification, in particular, of functional data. SVM is frequently used with a kernel (e.g.: Gaussian) which involves a scalar bandwidth parameter. In this paper, we pro- pose to use kernels with functional bandwidths. In this way, accuracy may be improved, and the time intervals critical for classification are identified. Tuning the functional parameters of the new kernel is a challenging task expressed as a continuous optimization problem, solved by means of a heuristic. Our experiments with benchmark data sets show the advantages of using functional parameters and the ef- fectiveness of our approach.ELSEVIER SCIENCE BVEstadística e Investigación OperativaFQM329: Optimización2018info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfapplication/pdfhttps://hdl.handle.net/11441/107638https://doi.org/10.1016/j.ejor.2018.11.024reponame:idUS. Depósito de Investigación de la Universidad de Sevillainstname:Universidad de Sevilla (US)InglésEuropean Journal of Operational Research, 275 (1), 195-207.http://doi.org/10.1016/j.ejor.2018.11.024info:eu-repo/semantics/openAccessoai:idus.us.es:11441/1076382026-06-17T12:51:07Z
dc.title.none.fl_str_mv Functional-bandwidth kernel for Support Vector Machine with Functional Data_An alternating optimization algorithm
title Functional-bandwidth kernel for Support Vector Machine with Functional Data_An alternating optimization algorithm
spellingShingle Functional-bandwidth kernel for Support Vector Machine with Functional Data_An alternating optimization algorithm
Blanquero Bravo, Rafael
Data mining
Functional Data classification
Parameter tuning
SVM
Functional bandwidth
title_short Functional-bandwidth kernel for Support Vector Machine with Functional Data_An alternating optimization algorithm
title_full Functional-bandwidth kernel for Support Vector Machine with Functional Data_An alternating optimization algorithm
title_fullStr Functional-bandwidth kernel for Support Vector Machine with Functional Data_An alternating optimization algorithm
title_full_unstemmed Functional-bandwidth kernel for Support Vector Machine with Functional Data_An alternating optimization algorithm
title_sort Functional-bandwidth kernel for Support Vector Machine with Functional Data_An alternating optimization algorithm
dc.creator.none.fl_str_mv Blanquero Bravo, Rafael
Carrizosa Priego, Emilio José
Jiménez Cordero, María Asunción
Martín Barragán, Belén
author Blanquero Bravo, Rafael
author_facet Blanquero Bravo, Rafael
Carrizosa Priego, Emilio José
Jiménez Cordero, María Asunción
Martín Barragán, Belén
author_role author
author2 Carrizosa Priego, Emilio José
Jiménez Cordero, María Asunción
Martín Barragán, Belén
author2_role author
author
author
dc.contributor.none.fl_str_mv Estadística e Investigación Operativa
FQM329: Optimización
dc.subject.none.fl_str_mv Data mining
Functional Data classification
Parameter tuning
SVM
Functional bandwidth
topic Data mining
Functional Data classification
Parameter tuning
SVM
Functional bandwidth
description Functional Data Analysis (FDA) is devoted to the study of data which are functions. Support Vector Ma- chine (SVM) is a benchmark tool for classification, in particular, of functional data. SVM is frequently used with a kernel (e.g.: Gaussian) which involves a scalar bandwidth parameter. In this paper, we pro- pose to use kernels with functional bandwidths. In this way, accuracy may be improved, and the time intervals critical for classification are identified. Tuning the functional parameters of the new kernel is a challenging task expressed as a continuous optimization problem, solved by means of a heuristic. Our experiments with benchmark data sets show the advantages of using functional parameters and the ef- fectiveness of our approach.
publishDate 2018
dc.date.none.fl_str_mv 2018
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv https://hdl.handle.net/11441/107638
https://doi.org/10.1016/j.ejor.2018.11.024
url https://hdl.handle.net/11441/107638
https://doi.org/10.1016/j.ejor.2018.11.024
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv European Journal of Operational Research, 275 (1), 195-207.
http://doi.org/10.1016/j.ejor.2018.11.024
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv ELSEVIER SCIENCE BV
publisher.none.fl_str_mv ELSEVIER SCIENCE BV
dc.source.none.fl_str_mv reponame:idUS. Depósito de Investigación de la Universidad de Sevilla
instname:Universidad de Sevilla (US)
instname_str Universidad de Sevilla (US)
reponame_str idUS. Depósito de Investigación de la Universidad de Sevilla
collection idUS. Depósito de Investigación de la Universidad de Sevilla
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869408407780327424
score 15.301629

Similar Items