Massive missing data reconstruction in ocean buoys with evolutionary product unit neural networks

In this paper we tackle the problem of massive missing data reconstruction in ocean buoys, with a Evolutionary Product Unit Neural Network (EPUNN). When considering a large number of buoys to reconstruct missing data, it is sometimes di cult to nd a common period of completeness (without missing dat...

ver descrição completa

Detalhes bibliográficos
Autores: Durán-Rosal, Antonio Manuel, Hervás Martínez, César, Tallón-Ballesteros, A.J., Martínez Estudillo, Alfonso Carlos, Salcedo-Sanz, S.
Formato: artículo
Fecha de publicación:2016
País:España
Recursos:Universidad Loyola Andalucía
Repositorio:Brújula
OAI Identifier:oai:dnet:brújula_____::b2d3716e2c92cbcba76d5006e3ecb336
Acesso em linha:https://hdl.handle.net/20.500.12412/7182
Access Level:acceso abierto
Palavra-chave:Significant wave height
Missing values reconstruction
Product Unit Neural Networks
Evolutionary Algorithm
id ES_2fa2fa6a23da87f43c923b972a29a0e8
oai_identifier_str oai:dnet:brújula_____::b2d3716e2c92cbcba76d5006e3ecb336
network_acronym_str ES
network_name_str España
repository_id_str
spelling Massive missing data reconstruction in ocean buoys with evolutionary product unit neural networksDurán-Rosal, Antonio ManuelHervás Martínez, CésarTallón-Ballesteros, A.J.Martínez Estudillo, Alfonso CarlosSalcedo-Sanz, S.Significant wave heightMissing values reconstructionProduct Unit Neural NetworksEvolutionary AlgorithmIn this paper we tackle the problem of massive missing data reconstruction in ocean buoys, with a Evolutionary Product Unit Neural Network (EPUNN). When considering a large number of buoys to reconstruct missing data, it is sometimes di cult to nd a common period of completeness (without missing data on it) in the data to form a proper training and test set. In this paper we solve this issue by using partial reconstruction, which are then used as inputs of the EPUNN, with linear models. Missing data reconstruction in several phases or steps is then proposed. In this work we also show the potential of EPUNN to obtain simple, interpretable models in spite of the non-linear characteristic of the network, much simpler than the commonly used sigmoid-based neural systems. In the experimental section of the paper we show the performance of the proposed approach in a real case of massive missing data reconstruction in 6 wave-rider buoys at the Gulf of Alaska.Es la versión aceptada del artículo. Se puede consultar la versión final en https://doi.org/10.1016/j.oceaneng.2016.03.053MICYT2016info:eu-repo/semantics/articlehttps://hdl.handle.net/20.500.12412/7182reponame:Brújulainstname:Universidad Loyola AndalucíaInglésTIN2014-54583-C2-1-Rhttps://doi.org/10.1016/j.oceaneng.2016.03.053info:eu-repo/semantics/openAccessoai:dnet:brújula_____::b2d3716e2c92cbcba76d5006e3ecb3362026-06-24T12:48:37Z
dc.title.none.fl_str_mv Massive missing data reconstruction in ocean buoys with evolutionary product unit neural networks
title Massive missing data reconstruction in ocean buoys with evolutionary product unit neural networks
spellingShingle Massive missing data reconstruction in ocean buoys with evolutionary product unit neural networks
Durán-Rosal, Antonio Manuel
Significant wave height
Missing values reconstruction
Product Unit Neural Networks
Evolutionary Algorithm
title_short Massive missing data reconstruction in ocean buoys with evolutionary product unit neural networks
title_full Massive missing data reconstruction in ocean buoys with evolutionary product unit neural networks
title_fullStr Massive missing data reconstruction in ocean buoys with evolutionary product unit neural networks
title_full_unstemmed Massive missing data reconstruction in ocean buoys with evolutionary product unit neural networks
title_sort Massive missing data reconstruction in ocean buoys with evolutionary product unit neural networks
dc.creator.none.fl_str_mv Durán-Rosal, Antonio Manuel
Hervás Martínez, César
Tallón-Ballesteros, A.J.
Martínez Estudillo, Alfonso Carlos
Salcedo-Sanz, S.
author Durán-Rosal, Antonio Manuel
author_facet Durán-Rosal, Antonio Manuel
Hervás Martínez, César
Tallón-Ballesteros, A.J.
Martínez Estudillo, Alfonso Carlos
Salcedo-Sanz, S.
author_role author
author2 Hervás Martínez, César
Tallón-Ballesteros, A.J.
Martínez Estudillo, Alfonso Carlos
Salcedo-Sanz, S.
author2_role author
author
author
author
dc.subject.none.fl_str_mv Significant wave height
Missing values reconstruction
Product Unit Neural Networks
Evolutionary Algorithm
topic Significant wave height
Missing values reconstruction
Product Unit Neural Networks
Evolutionary Algorithm
description In this paper we tackle the problem of massive missing data reconstruction in ocean buoys, with a Evolutionary Product Unit Neural Network (EPUNN). When considering a large number of buoys to reconstruct missing data, it is sometimes di cult to nd a common period of completeness (without missing data on it) in the data to form a proper training and test set. In this paper we solve this issue by using partial reconstruction, which are then used as inputs of the EPUNN, with linear models. Missing data reconstruction in several phases or steps is then proposed. In this work we also show the potential of EPUNN to obtain simple, interpretable models in spite of the non-linear characteristic of the network, much simpler than the commonly used sigmoid-based neural systems. In the experimental section of the paper we show the performance of the proposed approach in a real case of massive missing data reconstruction in 6 wave-rider buoys at the Gulf of Alaska.
publishDate 2016
dc.date.none.fl_str_mv 2016
dc.type.none.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://hdl.handle.net/20.500.12412/7182
url https://hdl.handle.net/20.500.12412/7182
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv TIN2014-54583-C2-1-R
https://doi.org/10.1016/j.oceaneng.2016.03.053
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.source.none.fl_str_mv reponame:Brújula
instname:Universidad Loyola Andalucía
instname_str Universidad Loyola Andalucía
reponame_str Brújula
collection Brújula
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869405491002605568
score 15,811543

Registos relacionados