Modeling and Forecasting Cases of RSV Using Artificial Neural Networks

In this paper, we study and present a mathematical modeling approach based on artificial neural networks to forecast the number of cases of respiratory syncytial virus (RSV). The number of RSV-positive cases in most of the countries around the world present a seasonal-type behavior. We constructed a...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Myladis R. Cogollo, Gilberto González-Parra, Abraham J. Arenas
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	forecasting epidemics respiratory syncytial virus mathematical modeling artificial neural networks seasonality Mathematics QA1-939
Acceso en línea:	https://doaj.org/article/119951f3937c4bfabc088e36a0195a36
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:119951f3937c4bfabc088e36a0195a36
record_format	dspace
spelling	oai:doaj.org-article:119951f3937c4bfabc088e36a0195a362021-11-25T18:17:34ZModeling and Forecasting Cases of RSV Using Artificial Neural Networks10.3390/math92229582227-7390https://doaj.org/article/119951f3937c4bfabc088e36a0195a362021-11-01T00:00:00Zhttps://www.mdpi.com/2227-7390/9/22/2958https://doaj.org/toc/2227-7390In this paper, we study and present a mathematical modeling approach based on artificial neural networks to forecast the number of cases of respiratory syncytial virus (RSV). The number of RSV-positive cases in most of the countries around the world present a seasonal-type behavior. We constructed and developed several multilayer perceptron (MLP) models that intend to appropriately forecast the number of cases of RSV, based on previous history. We compared our mathematical modeling approach with a classical statistical technique for the time-series, and we concluded that our results are more accurate. The dataset collected during 2005 to 2010 consisting of 312 weeks belongs to Bogotá D.C., Colombia. The adjusted MLP network that we constructed has a fairly high forecast accuracy. Finally, based on these computations, we recommend training the selected MLP model using 70% of the historical data of RSV-positive cases for training and 20% for validation in order to obtain more accurate results. These results are useful and provide scientific information for health authorities of Colombia to design suitable public health policies related to RSV.Myladis R. CogolloGilberto González-ParraAbraham J. ArenasMDPI AGarticleforecastingepidemicsrespiratory syncytial virusmathematical modelingartificial neural networksseasonalityMathematicsQA1-939ENMathematics, Vol 9, Iss 2958, p 2958 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	forecasting epidemics respiratory syncytial virus mathematical modeling artificial neural networks seasonality Mathematics QA1-939
spellingShingle	forecasting epidemics respiratory syncytial virus mathematical modeling artificial neural networks seasonality Mathematics QA1-939 Myladis R. Cogollo Gilberto González-Parra Abraham J. Arenas Modeling and Forecasting Cases of RSV Using Artificial Neural Networks
description	In this paper, we study and present a mathematical modeling approach based on artificial neural networks to forecast the number of cases of respiratory syncytial virus (RSV). The number of RSV-positive cases in most of the countries around the world present a seasonal-type behavior. We constructed and developed several multilayer perceptron (MLP) models that intend to appropriately forecast the number of cases of RSV, based on previous history. We compared our mathematical modeling approach with a classical statistical technique for the time-series, and we concluded that our results are more accurate. The dataset collected during 2005 to 2010 consisting of 312 weeks belongs to Bogotá D.C., Colombia. The adjusted MLP network that we constructed has a fairly high forecast accuracy. Finally, based on these computations, we recommend training the selected MLP model using 70% of the historical data of RSV-positive cases for training and 20% for validation in order to obtain more accurate results. These results are useful and provide scientific information for health authorities of Colombia to design suitable public health policies related to RSV.
format	article
author	Myladis R. Cogollo Gilberto González-Parra Abraham J. Arenas
author_facet	Myladis R. Cogollo Gilberto González-Parra Abraham J. Arenas
author_sort	Myladis R. Cogollo
title	Modeling and Forecasting Cases of RSV Using Artificial Neural Networks
title_short	Modeling and Forecasting Cases of RSV Using Artificial Neural Networks
title_full	Modeling and Forecasting Cases of RSV Using Artificial Neural Networks
title_fullStr	Modeling and Forecasting Cases of RSV Using Artificial Neural Networks
title_full_unstemmed	Modeling and Forecasting Cases of RSV Using Artificial Neural Networks
title_sort	modeling and forecasting cases of rsv using artificial neural networks
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/119951f3937c4bfabc088e36a0195a36
work_keys_str_mv	AT myladisrcogollo modelingandforecastingcasesofrsvusingartificialneuralnetworks AT gilbertogonzalezparra modelingandforecastingcasesofrsvusingartificialneuralnetworks AT abrahamjarenas modelingandforecastingcasesofrsvusingartificialneuralnetworks
_version_	1718411363789832192

Modeling and Forecasting Cases of RSV Using Artificial Neural Networks

Ejemplares similares