Selection of time instants and intervals with Support Vector Regression for multivariate functional data

When continuously monitoring processes over time, data is collected along a whole period, from which only certain time instants and certain time intervals may play a crucial role in the data analysis. We develop a method that addresses the problem of selecting a finite and small set of short interva...

Descripción completa

Detalles Bibliográficos
Autores: Blanquero Bravo, Rafael, Carrizosa Priego, Emilio José, Jiménez Cordero, María Asunción, Martín Barragán, Belén
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2020
País:España
Institución:Universidad de Sevilla (US)
Repositorio:idUS. Depósito de Investigación de la Universidad de Sevilla
OAI Identifier:oai:idus.us.es:11441/107836
Acceso en línea:https://hdl.handle.net/11441/107836
https://doi.org/10.1016/j.cor.2020.105050
Access Level:acceso abierto
Palabra clave:Machine learning
Functional regression
Support Vector Regression
Time interval selection
id ES_fa9d30e3eeaf564e98e7dcf66466e660
oai_identifier_str oai:idus.us.es:11441/107836
network_acronym_str ES
network_name_str España
repository_id_str
spelling Selection of time instants and intervals with Support Vector Regression for multivariate functional dataBlanquero Bravo, RafaelCarrizosa Priego, Emilio JoséJiménez Cordero, María AsunciónMartín Barragán, BelénMachine learningFunctional regressionSupport Vector RegressionTime interval selectionWhen continuously monitoring processes over time, data is collected along a whole period, from which only certain time instants and certain time intervals may play a crucial role in the data analysis. We develop a method that addresses the problem of selecting a finite and small set of short intervals (or instants) able to capture the information needed to predict a response variable from multivariate functional data using Support Vector Regression (SVR). In addition to improving interpretability, storage requirements, and monitoring cost, feature selection can potentially reduce overfitting by mitigating data autocorrelation. We propose a continuous optimization algorithm to fit the SVR parameters and select intervals and instants. Our approach takes advantage of the functional nature of the data by formulating a new bilevel optimization problem that integrates selection of intervals and instants, tuning of some key SVR parameters and fitting the SVR. We illustrate the usefulness of our proposal in some benchmark data sets.PERGAMON-ELSEVIER SCIENCE LTDEstadística e Investigación OperativaFQM329: Optimización2020info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfapplication/pdfhttps://hdl.handle.net/11441/107836https://doi.org/10.1016/j.cor.2020.105050reponame:idUS. Depósito de Investigación de la Universidad de Sevillainstname:Universidad de Sevilla (US)InglésComputers & Operations Research, 123, 105050 - 1-105050 - 24.http://doi.org/10.1016/j.cor.2020.105050info:eu-repo/semantics/openAccessoai:idus.us.es:11441/1078362026-06-17T12:51:07Z
dc.title.none.fl_str_mv Selection of time instants and intervals with Support Vector Regression for multivariate functional data
title Selection of time instants and intervals with Support Vector Regression for multivariate functional data
spellingShingle Selection of time instants and intervals with Support Vector Regression for multivariate functional data
Blanquero Bravo, Rafael
Machine learning
Functional regression
Support Vector Regression
Time interval selection
title_short Selection of time instants and intervals with Support Vector Regression for multivariate functional data
title_full Selection of time instants and intervals with Support Vector Regression for multivariate functional data
title_fullStr Selection of time instants and intervals with Support Vector Regression for multivariate functional data
title_full_unstemmed Selection of time instants and intervals with Support Vector Regression for multivariate functional data
title_sort Selection of time instants and intervals with Support Vector Regression for multivariate functional data
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 Machine learning
Functional regression
Support Vector Regression
Time interval selection
topic Machine learning
Functional regression
Support Vector Regression
Time interval selection
description When continuously monitoring processes over time, data is collected along a whole period, from which only certain time instants and certain time intervals may play a crucial role in the data analysis. We develop a method that addresses the problem of selecting a finite and small set of short intervals (or instants) able to capture the information needed to predict a response variable from multivariate functional data using Support Vector Regression (SVR). In addition to improving interpretability, storage requirements, and monitoring cost, feature selection can potentially reduce overfitting by mitigating data autocorrelation. We propose a continuous optimization algorithm to fit the SVR parameters and select intervals and instants. Our approach takes advantage of the functional nature of the data by formulating a new bilevel optimization problem that integrates selection of intervals and instants, tuning of some key SVR parameters and fitting the SVR. We illustrate the usefulness of our proposal in some benchmark data sets.
publishDate 2020
dc.date.none.fl_str_mv 2020
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/11441/107836
https://doi.org/10.1016/j.cor.2020.105050
url https://hdl.handle.net/11441/107836
https://doi.org/10.1016/j.cor.2020.105050
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Computers & Operations Research, 123, 105050 - 1-105050 - 24.
http://doi.org/10.1016/j.cor.2020.105050
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 PERGAMON-ELSEVIER SCIENCE LTD
publisher.none.fl_str_mv PERGAMON-ELSEVIER SCIENCE LTD
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_ 1869425227193122816
score 15,300719

Similares en LA Referencia