Rainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting

Rainfall forecasting has gained utmost research relevance in recent times due to its complexities and persistent applications such as flood forecasting and monitoring of pollutant concentration levels, among others. Existing models use complex statistical models that are often too costly, both compu...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Ari Yair Barrera-Animas, Lukumon O. Oyedele, Muhammad Bilal, Taofeek Dolapo Akinosho, Juan Manuel Davila Delgado, Lukman Adewale Akanbi
Formato:	article
Lenguaje:	EN
Publicado:	Elsevier 2022
Materias:	Rainfall prediction LSTM Networks Multivariate time-series Multi-step forecast Time-series data Cybernetics Q300-390 Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/1a13908ff72e44c0bc20b21c4318b446
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:1a13908ff72e44c0bc20b21c4318b446
record_format	dspace
spelling	oai:doaj.org-article:1a13908ff72e44c0bc20b21c4318b4462021-11-26T04:41:56ZRainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting2666-827010.1016/j.mlwa.2021.100204https://doaj.org/article/1a13908ff72e44c0bc20b21c4318b4462022-03-01T00:00:00Zhttp://www.sciencedirect.com/science/article/pii/S266682702100102Xhttps://doaj.org/toc/2666-8270Rainfall forecasting has gained utmost research relevance in recent times due to its complexities and persistent applications such as flood forecasting and monitoring of pollutant concentration levels, among others. Existing models use complex statistical models that are often too costly, both computationally and budgetary, or are not applied to downstream applications. Therefore, approaches that use Machine Learning algorithms in conjunction with time-series data are being explored as an alternative to overcome these drawbacks. To this end, this study presents a comparative analysis using simplified rainfall estimation models based on conventional Machine Learning algorithms and Deep Learning architectures that are efficient for these downstream applications. Models based on LSTM, Stacked-LSTM, Bidirectional-LSTM Networks, XGBoost, and an ensemble of Gradient Boosting Regressor, Linear Support Vector Regression, and an Extra-trees Regressor were compared in the task of forecasting hourly rainfall volumes using time-series data. Climate data from 2000 to 2020 from five major cities in the United Kingdom were used. The evaluation metrics of Loss, Root Mean Squared Error, Mean Absolute Error, and Root Mean Squared Logarithmic Error were used to evaluate the models’ performance. Results show that a Bidirectional-LSTM Network can be used as a rainfall forecast model with comparable performance to Stacked-LSTM Networks. Among all the models tested, the Stacked-LSTM Network with two hidden layers and the Bidirectional-LSTM Network performed best. This suggests that models based on LSTM-Networks with fewer hidden layers perform better for this approach; denoting its ability to be applied as an approach for budget-wise rainfall forecast applications.Ari Yair Barrera-AnimasLukumon O. OyedeleMuhammad BilalTaofeek Dolapo AkinoshoJuan Manuel Davila DelgadoLukman Adewale AkanbiElsevierarticleRainfall predictionLSTM NetworksMultivariate time-seriesMulti-step forecastTime-series dataCyberneticsQ300-390Electronic computers. Computer scienceQA75.5-76.95ENMachine Learning with Applications, Vol 7, Iss , Pp 100204- (2022)
institution	DOAJ
collection	DOAJ
language	EN
topic	Rainfall prediction LSTM Networks Multivariate time-series Multi-step forecast Time-series data Cybernetics Q300-390 Electronic computers. Computer science QA75.5-76.95
spellingShingle	Rainfall prediction LSTM Networks Multivariate time-series Multi-step forecast Time-series data Cybernetics Q300-390 Electronic computers. Computer science QA75.5-76.95 Ari Yair Barrera-Animas Lukumon O. Oyedele Muhammad Bilal Taofeek Dolapo Akinosho Juan Manuel Davila Delgado Lukman Adewale Akanbi Rainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting
description	Rainfall forecasting has gained utmost research relevance in recent times due to its complexities and persistent applications such as flood forecasting and monitoring of pollutant concentration levels, among others. Existing models use complex statistical models that are often too costly, both computationally and budgetary, or are not applied to downstream applications. Therefore, approaches that use Machine Learning algorithms in conjunction with time-series data are being explored as an alternative to overcome these drawbacks. To this end, this study presents a comparative analysis using simplified rainfall estimation models based on conventional Machine Learning algorithms and Deep Learning architectures that are efficient for these downstream applications. Models based on LSTM, Stacked-LSTM, Bidirectional-LSTM Networks, XGBoost, and an ensemble of Gradient Boosting Regressor, Linear Support Vector Regression, and an Extra-trees Regressor were compared in the task of forecasting hourly rainfall volumes using time-series data. Climate data from 2000 to 2020 from five major cities in the United Kingdom were used. The evaluation metrics of Loss, Root Mean Squared Error, Mean Absolute Error, and Root Mean Squared Logarithmic Error were used to evaluate the models’ performance. Results show that a Bidirectional-LSTM Network can be used as a rainfall forecast model with comparable performance to Stacked-LSTM Networks. Among all the models tested, the Stacked-LSTM Network with two hidden layers and the Bidirectional-LSTM Network performed best. This suggests that models based on LSTM-Networks with fewer hidden layers perform better for this approach; denoting its ability to be applied as an approach for budget-wise rainfall forecast applications.
format	article
author	Ari Yair Barrera-Animas Lukumon O. Oyedele Muhammad Bilal Taofeek Dolapo Akinosho Juan Manuel Davila Delgado Lukman Adewale Akanbi
author_facet	Ari Yair Barrera-Animas Lukumon O. Oyedele Muhammad Bilal Taofeek Dolapo Akinosho Juan Manuel Davila Delgado Lukman Adewale Akanbi
author_sort	Ari Yair Barrera-Animas
title	Rainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting
title_short	Rainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting
title_full	Rainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting
title_fullStr	Rainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting
title_full_unstemmed	Rainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting
title_sort	rainfall prediction: a comparative analysis of modern machine learning algorithms for time-series forecasting
publisher	Elsevier
publishDate	2022
url	https://doaj.org/article/1a13908ff72e44c0bc20b21c4318b446
work_keys_str_mv	AT ariyairbarreraanimas rainfallpredictionacomparativeanalysisofmodernmachinelearningalgorithmsfortimeseriesforecasting AT lukumonooyedele rainfallpredictionacomparativeanalysisofmodernmachinelearningalgorithmsfortimeseriesforecasting AT muhammadbilal rainfallpredictionacomparativeanalysisofmodernmachinelearningalgorithmsfortimeseriesforecasting AT taofeekdolapoakinosho rainfallpredictionacomparativeanalysisofmodernmachinelearningalgorithmsfortimeseriesforecasting AT juanmanueldaviladelgado rainfallpredictionacomparativeanalysisofmodernmachinelearningalgorithmsfortimeseriesforecasting AT lukmanadewaleakanbi rainfallpredictionacomparativeanalysisofmodernmachinelearningalgorithmsfortimeseriesforecasting
_version_	1718409764466065408

Rainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting

Ejemplares similares