Comparative analysis of Gated Recurrent Units (GRU), long Short-Term memory (LSTM) cells, autoregressive Integrated moving average (ARIMA), seasonal autoregressive Integrated moving average (SARIMA) for forecasting COVID-19 trends

ArunKumar, K.E.; Kalaga, Dinesh V.; Kumar, Ch. Mohan Sai; Kawaji, Masahiro; Brenza, Timothy M.

doi:10.1016/j.aej.2022.01.011

Cited by 133 publications

(72 citation statements)

References 37 publications

(36 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this study, two different deep learning models based on GRU and LSTM, which make productivity estimation on wheat data of Konya province, are proposed to meet the stated requirement. Since the two proposed deep learning models are based on RNN [41][42][43][44], both the performance and training times of the models were compared. As a result of the comparison processes, it is seen that the results of the LSTM model are slightly better than the GRU model.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

Buğday Veri̇m Tahmi̇ni̇ İçi̇n Yeni̇lemeli̇ Si̇ni̇r Aği Tabanli Model Geli̇şti̇rme

ÇETİNER

Kara

2022

Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi

View full text Add to dashboard Cite

In the study carried out in line with the stated purposes, monthly rain, humidity and temperature data, wheat production amount, and wheat productivity data of Konya province between 1980-2020 were used. Using these data, wheat productivity estimation was performed with (Gated Recurrent Units) GRU and Long Short Term Memory (LSTM) methods, which are Recurrent Neural Network (RNN) based algorithms. When wheat productivity estimation performance was examined with the implemented GRU-based model, 0.9550, 0.0059, 0.0280, 0.0623, 7.45 values were obtained for the R 2 score, MSE, RMSE, MAE and MAPE values, respectively. In the performance results obtained with the LSTM method, which is another RNN-based method, 0.9667, 0.0054, 0.0280, 0.0614, 7.33 values were obtained for the R 2 score, MSE, RMSE, MAE and MAPE values, respectively. Although the LSTM method gave better results than the GRU method, the training modelling time of the LSTM method took longer than that of the GRU method.

show abstract

Section: Conclusion and Discussionmentioning

confidence: 99%

Buğday Veri̇m Tahmi̇ni̇ İçi̇n Yeni̇lemeli̇ Si̇ni̇r Aği Tabanli Model Geli̇şti̇rme

ÇETİNER

Kara

2022

Adıyaman Üniversitesi Mühendislik Bilimleri Dergisi

View full text Add to dashboard Cite

show abstract

“… Egypt, Kuwait; May - Dec, 2020 0.47, 0.73. Arun et al, 2022 [8] Compared GRU and LSTM vs ARIMA and SARIMA models and found GRU and LSTM to achieve the lowest error in most cases. Top 10 countries; Jan, 2020 - Jun, 2021 RMSE 8K-25K.…”

Section: Deep Learning Models For Covid-19 Forecastingmentioning

confidence: 99%

“…To provide a more in-depth view of the existing RNN-based forecasting models, we focus on five representative studies: Arun et al [8] , Dairi et al [18] , Ma et al [52] , Omran et al [65] , and Pavlyutin et al [67] …”

Section: Deep Learning Models For Covid-19 Forecastingmentioning

confidence: 99%

Deep learning for Covid-19 forecasting: State-of-the-art review.

Kamalov¹,

Rajab²,

Cherukuri³

et al. 2022

Neurocomputing

View full text Add to dashboard Cite

“…But consumes much time. [24] Comparative analysis of Gated Recurrent Units (GRU), long Short-Term memory (LSTM) cells, autoregressive Integrated moving average (ARIMA), seasonal autoregressive Integrated moving average (SARIMA) for forecasting COVID-19 trend SARIMA based models performed better for India, Russia, Peru, Chile, and the UK and ARIMA model performed better for Brazil. For Mexico and Iran, LSTM model performed better and the GRU model performed better.…”

Section: Literature Review On Modelling and Forcasting Covid-19mentioning

confidence: 99%

Attributable Fraction and Forecasting for COVID-19 Confirmed Cases in Nigeria Using Facebook- Prophet Machine Learning Model

Ibidoja

Fowobaje

2022

AJPAS

View full text Add to dashboard Cite

Aims: The motivation is to know the attributable fraction among Nigerians who tested positive for covid-19 and forecast the covid-19 cases. Place and Duration of Study: We extracted data from (https://covid19.ncdc.gov.ng/) on 8th September,2021 and covid.19analytics package on 7th September, 2021, from Data Repository by Johns Hopkins University Center for Systems Science and Engineering , Status of Cases in Toronto – City of Toronto , COVID-19: Open Data Toronto ,COVID-19: Health Canada , Severe acute respiratory syndrome coronavirus 2 isolate Wuhan-Hu-1, COViD-19 Vaccination and Testing records from “Our World In Data” and Pandemics historical records from Visual Capitalist. Data in Nigeria contained the number of samples tested, confirmed cases, active cases, discharged cases and deaths. Methodology: Attributable fraction was used to compute the proportion of patients who tested positive to Covid-19. By using the time for regressor, Prophet model will fit many non-linear and linear functions of time components. Prophet uses the Fourier series to get flexible model to forecast and fit the seasonality effects. A fast solution for L-BFGS which stands for Limited memory Broyden-Fletcher-Goldfarb-Shannon algorithm, is used with Stan backend for the prediction problem. Results: As at Saturday 11th September 2021,7:18am Nigeria local time, a total of 2884034 samples have been tested for covid-19, with 198239 confirmed cases,9871 active cases,185780 discharged cases and 2588 deaths. The attributable fraction for covid-19 in Nigeria was 0.0687. The r square is very high (0.999), and the p value is very low (2.2e-16). Conclusion: The attributable fraction gives the percentage of the patients who tested positive to covid-19, among the 2884034 samples tested. It implies that the remaining percentage of patients who tested negative to covid-19 only exhibit covid-19 symptoms or were exposed to the virus. The confirmed cases were found to be highest on Saturdays with the lowest on Tuesdays.

show abstract

Comparative analysis of Gated Recurrent Units (GRU), long Short-Term memory (LSTM) cells, autoregressive Integrated moving average (ARIMA), seasonal autoregressive Integrated moving average (SARIMA) for forecasting COVID-19 trends

Cited by 133 publications

References 37 publications

Buğday Veri̇m Tahmi̇ni̇ İçi̇n Yeni̇lemeli̇ Si̇ni̇r Aği Tabanli Model Geli̇şti̇rme

Buğday Veri̇m Tahmi̇ni̇ İçi̇n Yeni̇lemeli̇ Si̇ni̇r Aği Tabanli Model Geli̇şti̇rme

Deep learning for Covid-19 forecasting: State-of-the-art review.

Attributable Fraction and Forecasting for COVID-19 Confirmed Cases in Nigeria Using Facebook- Prophet Machine Learning Model

Contact Info

Product

Resources

About