PHILNet: A novel efficient approach for time series forecasting using deep learning

Time series is one of the most common data types in the industry nowadays. Forecasting the future of a time series behavior can be useful in planning ahead, saving time, resources, and helping avoid undesired scenarios. To make the forecasting, historical data is utilized due to the causal nature of...

ver descrição completa

Detalhes bibliográficos
Autores: Jiménez Navarro, Manuel Jesús, Martínez Ballesteros, María, Martínez-Álvarez, Francisco, Asencio Cortés, Gualberto
Tipo de documento: artigo
Data de publicação:2023
País:España
Recursos:Universidad Pablo de Olavide (UPO)
Repositório:RIO. Repositorio Institucional Olavide
Idioma:inglês
OAI Identifier:oai:rio.upo.es:10433/19782
Acesso em linha:https://hdl.handle.net/10433/19782
Access Level:Acceso aberto
Palavra-chave:Time series
Forecasting
Deep learning
Efficiency
id ES_2dd5dadb8fc86afebba30f95f104d0a3
oai_identifier_str oai:rio.upo.es:10433/19782
network_acronym_str ES
network_name_str España
repository_id_str
spelling PHILNet: A novel efficient approach for time series forecasting using deep learningJiménez Navarro, Manuel JesúsMartínez Ballesteros, MaríaMartínez-Álvarez, FranciscoAsencio Cortés, GualbertoTime seriesForecastingDeep learningEfficiencyTime series is one of the most common data types in the industry nowadays. Forecasting the future of a time series behavior can be useful in planning ahead, saving time, resources, and helping avoid undesired scenarios. To make the forecasting, historical data is utilized due to the causal nature of the time series. Several deep learning algorithms have been presented in this area, where the input is processed through a series of non-linear functions to produce the output. We present a novel strategy to improve the performance of deep learning models in time series forecasting in terms of efficiency while reaching similar effectiveness. This approach separates the model into levels, starting with the easiest and continuing to the most difficult. The simpler levels deal with smoothed versions of the input, whereas the most sophisticated level deals with the raw data. This strategy seeks to mimic the human learning process, in which basic tasks are completed initially, followed by more precise and sophisticated ones. Our method achieved promising results, obtaining a 35% improvement in mean squared error and a 2.6 time decrease in training time compared with the best models found in a variety of time series.Elsevier20242024-02-0620232023-03-1520232023-03-15journal articlehttp://purl.org/coar/resource_type/c_6501info:eu-repo/semantics/articleapplication/pdfhttps://hdl.handle.net/10433/19782reponame:RIO. Repositorio Institucional Olavideinstname:Universidad Pablo de Olavide (UPO)Inglésengopen accesshttp://purl.org/coar/access_right/c_abf2Attribution-ShareAlike 4.0 Internationalhttp://creativecommons.org/licenses/by-sa/4.0/info:eu-repo/semantics/openAccessoai:rio.upo.es:10433/197822026-06-13T12:46:27Z
dc.title.none.fl_str_mv PHILNet: A novel efficient approach for time series forecasting using deep learning
title PHILNet: A novel efficient approach for time series forecasting using deep learning
spellingShingle PHILNet: A novel efficient approach for time series forecasting using deep learning
Jiménez Navarro, Manuel Jesús
Time series
Forecasting
Deep learning
Efficiency
title_short PHILNet: A novel efficient approach for time series forecasting using deep learning
title_full PHILNet: A novel efficient approach for time series forecasting using deep learning
title_fullStr PHILNet: A novel efficient approach for time series forecasting using deep learning
title_full_unstemmed PHILNet: A novel efficient approach for time series forecasting using deep learning
title_sort PHILNet: A novel efficient approach for time series forecasting using deep learning
dc.creator.none.fl_str_mv Jiménez Navarro, Manuel Jesús
Martínez Ballesteros, María
Martínez-Álvarez, Francisco
Asencio Cortés, Gualberto
author Jiménez Navarro, Manuel Jesús
author_facet Jiménez Navarro, Manuel Jesús
Martínez Ballesteros, María
Martínez-Álvarez, Francisco
Asencio Cortés, Gualberto
author_role author
author2 Martínez Ballesteros, María
Martínez-Álvarez, Francisco
Asencio Cortés, Gualberto
author2_role author
author
author
dc.contributor.none.fl_str_mv
dc.subject.none.fl_str_mv Time series
Forecasting
Deep learning
Efficiency
topic Time series
Forecasting
Deep learning
Efficiency
description Time series is one of the most common data types in the industry nowadays. Forecasting the future of a time series behavior can be useful in planning ahead, saving time, resources, and helping avoid undesired scenarios. To make the forecasting, historical data is utilized due to the causal nature of the time series. Several deep learning algorithms have been presented in this area, where the input is processed through a series of non-linear functions to produce the output. We present a novel strategy to improve the performance of deep learning models in time series forecasting in terms of efficiency while reaching similar effectiveness. This approach separates the model into levels, starting with the easiest and continuing to the most difficult. The simpler levels deal with smoothed versions of the input, whereas the most sophisticated level deals with the raw data. This strategy seeks to mimic the human learning process, in which basic tasks are completed initially, followed by more precise and sophisticated ones. Our method achieved promising results, obtaining a 35% improvement in mean squared error and a 2.6 time decrease in training time compared with the best models found in a variety of time series.
publishDate 2023
dc.date.none.fl_str_mv 2023
2023-03-15
2023
2023-03-15
2024
2024-02-06
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://hdl.handle.net/10433/19782
url https://hdl.handle.net/10433/19782
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
Attribution-ShareAlike 4.0 International
http://creativecommons.org/licenses/by-sa/4.0/
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
Attribution-ShareAlike 4.0 International
http://creativecommons.org/licenses/by-sa/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:RIO. Repositorio Institucional Olavide
instname:Universidad Pablo de Olavide (UPO)
instname_str Universidad Pablo de Olavide (UPO)
reponame_str RIO. Repositorio Institucional Olavide
collection RIO. Repositorio Institucional Olavide
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869405357079527424
score 15.301603

Registros relacionados