Cargando…

A novel cross-validation strategy for artificial neural networks using distributed-lag environmental factors

In recent years, machine learning methods have been applied to various prediction scenarios in time-series data. However, some processing procedures such as cross-validation (CV) that rearrange the order of the longitudinal data might ruin the seriality and lead to a potentially biased outcome. Rega...

Descripción completa

Detalles Bibliográficos
Autores principales: Guo, Chao-Yu, Liu, Tse-Wei, Chen, Yi-Hau
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Public Library of Science 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7790373/
https://www.ncbi.nlm.nih.gov/pubmed/33411794
http://dx.doi.org/10.1371/journal.pone.0244094
_version_ 1783633410536767488
author Guo, Chao-Yu
Liu, Tse-Wei
Chen, Yi-Hau
author_facet Guo, Chao-Yu
Liu, Tse-Wei
Chen, Yi-Hau
author_sort Guo, Chao-Yu
collection PubMed
description In recent years, machine learning methods have been applied to various prediction scenarios in time-series data. However, some processing procedures such as cross-validation (CV) that rearrange the order of the longitudinal data might ruin the seriality and lead to a potentially biased outcome. Regarding this issue, a recent study investigated how different types of CV methods influence the predictive errors in conventional time-series data. Here, we examine a more complex distributed lag nonlinear model (DLNM), which has been widely used to assess the cumulative impacts of past exposures on the current health outcome. This research extends the DLNM into an artificial neural network (ANN) and investigates how the ANN model reacts to various CV schemes that result in different predictive biases. We also propose a newly designed permutation ratio to evaluate the performance of the CV in the ANN. This ratio mimics the concept of the R-square in conventional statistical regression models. The results show that as the complexity of the ANN increases, the predicted outcome becomes more stable, and the bias shows a decreasing trend. Among the different settings of hyperparameters, the novel strategy, Leave One Block Out Cross-Validation (LOBO-CV), demonstrated much better results, and the lowest mean square error was observed. The hyperparameters of the ANN trained by the LOBO-CV yielded the minimum number of prediction errors. The newly proposed permutation ratio indicates that LOBO-CV can contribute up to 34% of the prediction accuracy.
format Online
Article
Text
id pubmed-7790373
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher Public Library of Science
record_format MEDLINE/PubMed
spelling pubmed-77903732021-01-27 A novel cross-validation strategy for artificial neural networks using distributed-lag environmental factors Guo, Chao-Yu Liu, Tse-Wei Chen, Yi-Hau PLoS One Research Article In recent years, machine learning methods have been applied to various prediction scenarios in time-series data. However, some processing procedures such as cross-validation (CV) that rearrange the order of the longitudinal data might ruin the seriality and lead to a potentially biased outcome. Regarding this issue, a recent study investigated how different types of CV methods influence the predictive errors in conventional time-series data. Here, we examine a more complex distributed lag nonlinear model (DLNM), which has been widely used to assess the cumulative impacts of past exposures on the current health outcome. This research extends the DLNM into an artificial neural network (ANN) and investigates how the ANN model reacts to various CV schemes that result in different predictive biases. We also propose a newly designed permutation ratio to evaluate the performance of the CV in the ANN. This ratio mimics the concept of the R-square in conventional statistical regression models. The results show that as the complexity of the ANN increases, the predicted outcome becomes more stable, and the bias shows a decreasing trend. Among the different settings of hyperparameters, the novel strategy, Leave One Block Out Cross-Validation (LOBO-CV), demonstrated much better results, and the lowest mean square error was observed. The hyperparameters of the ANN trained by the LOBO-CV yielded the minimum number of prediction errors. The newly proposed permutation ratio indicates that LOBO-CV can contribute up to 34% of the prediction accuracy. Public Library of Science 2021-01-07 /pmc/articles/PMC7790373/ /pubmed/33411794 http://dx.doi.org/10.1371/journal.pone.0244094 Text en © 2021 Guo et al http://creativecommons.org/licenses/by/4.0/ This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
spellingShingle Research Article
Guo, Chao-Yu
Liu, Tse-Wei
Chen, Yi-Hau
A novel cross-validation strategy for artificial neural networks using distributed-lag environmental factors
title A novel cross-validation strategy for artificial neural networks using distributed-lag environmental factors
title_full A novel cross-validation strategy for artificial neural networks using distributed-lag environmental factors
title_fullStr A novel cross-validation strategy for artificial neural networks using distributed-lag environmental factors
title_full_unstemmed A novel cross-validation strategy for artificial neural networks using distributed-lag environmental factors
title_short A novel cross-validation strategy for artificial neural networks using distributed-lag environmental factors
title_sort novel cross-validation strategy for artificial neural networks using distributed-lag environmental factors
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7790373/
https://www.ncbi.nlm.nih.gov/pubmed/33411794
http://dx.doi.org/10.1371/journal.pone.0244094
work_keys_str_mv AT guochaoyu anovelcrossvalidationstrategyforartificialneuralnetworksusingdistributedlagenvironmentalfactors
AT liutsewei anovelcrossvalidationstrategyforartificialneuralnetworksusingdistributedlagenvironmentalfactors
AT chenyihau anovelcrossvalidationstrategyforartificialneuralnetworksusingdistributedlagenvironmentalfactors
AT guochaoyu novelcrossvalidationstrategyforartificialneuralnetworksusingdistributedlagenvironmentalfactors
AT liutsewei novelcrossvalidationstrategyforartificialneuralnetworksusingdistributedlagenvironmentalfactors
AT chenyihau novelcrossvalidationstrategyforartificialneuralnetworksusingdistributedlagenvironmentalfactors