A novel distributed forecasting method based on information fusion and incremental learning for streaming time series

Real-time algorithms have to adapt and adjust to new incoming patterns to provide timely and accurate responses. This paper presents a new distributed forecasting algorithm for streaming time series called StreamWNN. StreamWNN starts with an offline stage in which a forecasting model based on tuples...

ver descrição completa

Detalhes bibliográficos
Autores: Merlgar García, Laura, Gutiérrez Avilés, David, Rubio Escudero, Cristina, Troncoso, Alicia
Tipo de documento: artigo
Estado:Versão publicada
Data de publicação:2023
País:España
Recursos:Universidad de Sevilla (US)
Repositório:idUS. Depósito de Investigación de la Universidad de Sevilla
OAI Identifier:oai:idus.us.es:11441/166997
Acesso em linha:https://hdl.handle.net/11441/166997
https://doi.org/10.1016/j.inffus.2023.02.023
Access Level:Acceso aberto
Palavra-chave:Real-time forecasting
Incremental learning
Streaming time series
Electricity demand
id ES_e0bd6bf23e9180a8c977f0fa60a80f77
oai_identifier_str oai:idus.us.es:11441/166997
network_acronym_str ES
network_name_str España
repository_id_str
spelling A novel distributed forecasting method based on information fusion and incremental learning for streaming time seriesMerlgar García, LauraGutiérrez Avilés, DavidRubio Escudero, CristinaTroncoso, AliciaReal-time forecastingIncremental learningStreaming time seriesElectricity demandReal-time algorithms have to adapt and adjust to new incoming patterns to provide timely and accurate responses. This paper presents a new distributed forecasting algorithm for streaming time series called StreamWNN. StreamWNN starts with an offline stage in which a forecasting model based on tuples of information fusion is created with historical data. In particular, this model consists of the fusion of patterns composed of past values of the time series with the future values of their k-nearest neighbors. Afterwards, streaming data starts to arrive. The model is incrementally updated in the online stage using a buffer with streaming data that more accurately matches the current model patterns. The model can be updated daily, monthly, quarterly or based on error thresholds. The methodology has been applied to Spanish electricity demand time series providing more accurate results when the model is updated incrementally. The best error results are obtained with the daily update of the model, resulting in an error between 2% and 3.5% depending on the prediction horizon. The model provides better error results than other algorithms.ElsevierLenguajes y Sistemas InformáticosMinisterio de Ciencia e InnovaciónJunta de Andalucía2023info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfapplication/pdfhttps://hdl.handle.net/11441/166997https://doi.org/10.1016/j.inffus.2023.02.023reponame:idUS. Depósito de Investigación de la Universidad de Sevillainstname:Universidad de Sevilla (US)InglésInformation Fusion, 95, 163-173.PID2020-117954RBC21TED2021-131311B-C22PY20-00870UPO-138516https://www.sciencedirect.com/science/article/pii/S1566253523000635?via%3Dihubinfo:eu-repo/semantics/openAccessoai:idus.us.es:11441/1669972026-06-17T12:51:07Z
dc.title.none.fl_str_mv A novel distributed forecasting method based on information fusion and incremental learning for streaming time series
title A novel distributed forecasting method based on information fusion and incremental learning for streaming time series
spellingShingle A novel distributed forecasting method based on information fusion and incremental learning for streaming time series
Merlgar García, Laura
Real-time forecasting
Incremental learning
Streaming time series
Electricity demand
title_short A novel distributed forecasting method based on information fusion and incremental learning for streaming time series
title_full A novel distributed forecasting method based on information fusion and incremental learning for streaming time series
title_fullStr A novel distributed forecasting method based on information fusion and incremental learning for streaming time series
title_full_unstemmed A novel distributed forecasting method based on information fusion and incremental learning for streaming time series
title_sort A novel distributed forecasting method based on information fusion and incremental learning for streaming time series
dc.creator.none.fl_str_mv Merlgar García, Laura
Gutiérrez Avilés, David
Rubio Escudero, Cristina
Troncoso, Alicia
author Merlgar García, Laura
author_facet Merlgar García, Laura
Gutiérrez Avilés, David
Rubio Escudero, Cristina
Troncoso, Alicia
author_role author
author2 Gutiérrez Avilés, David
Rubio Escudero, Cristina
Troncoso, Alicia
author2_role author
author
author
dc.contributor.none.fl_str_mv Lenguajes y Sistemas Informáticos
Ministerio de Ciencia e Innovación
Junta de Andalucía
dc.subject.none.fl_str_mv Real-time forecasting
Incremental learning
Streaming time series
Electricity demand
topic Real-time forecasting
Incremental learning
Streaming time series
Electricity demand
description Real-time algorithms have to adapt and adjust to new incoming patterns to provide timely and accurate responses. This paper presents a new distributed forecasting algorithm for streaming time series called StreamWNN. StreamWNN starts with an offline stage in which a forecasting model based on tuples of information fusion is created with historical data. In particular, this model consists of the fusion of patterns composed of past values of the time series with the future values of their k-nearest neighbors. Afterwards, streaming data starts to arrive. The model is incrementally updated in the online stage using a buffer with streaming data that more accurately matches the current model patterns. The model can be updated daily, monthly, quarterly or based on error thresholds. The methodology has been applied to Spanish electricity demand time series providing more accurate results when the model is updated incrementally. The best error results are obtained with the daily update of the model, resulting in an error between 2% and 3.5% depending on the prediction horizon. The model provides better error results than other algorithms.
publishDate 2023
dc.date.none.fl_str_mv 2023
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/11441/166997
https://doi.org/10.1016/j.inffus.2023.02.023
url https://hdl.handle.net/11441/166997
https://doi.org/10.1016/j.inffus.2023.02.023
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Information Fusion, 95, 163-173.
PID2020-117954RBC21
TED2021-131311B-C22
PY20-00870
UPO-138516
https://www.sciencedirect.com/science/article/pii/S1566253523000635?via%3Dihub
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
publisher.none.fl_str_mv Elsevier
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_ 1869422229952921601
score 15,812455

Registros relacionados