A review on COVID-19 forecasting models

Neural Computing and Applications, Journal Year: 2021, Volume and Issue: 35(33), P. 23671 - 23681

Published: Feb. 4, 2021

The novel coronavirus (COVID-19) has spread to more than 200 countries worldwide, leading 36 million confirmed cases as of October 10, 2020. As such, several machine learning models that can forecast the outbreak globally have been released. This work presents a review and brief analysis most important forecasting against COVID-19. presented in this study possesses two parts. In first section, detailed scientometric an influential tool for bibliometric analyses, which were performed on COVID-19 data from Scopus Web Science databases. For above-mentioned analysis, keywords subject areas are addressed, while classification models, criteria evaluation, comparison solution approaches discussed second section work. conclusion discussion provided final sections study.

Language: Английский

---

Forecasting of COVID-19 cases using deep learning models: Is it reliable and practically significant?

Results in Physics, Journal Year: 2021, Volume and Issue: 21, P. 103817 - 103817

Published: Jan. 14, 2021

The ongoing outbreak of the COVID-19 pandemic prevails as an ultimatum to global economic growth and henceforth, all society since neither a curing drug nor preventing vaccine is discovered. spread increasing day by day, imposing human lives economy at risk. Due increased enormity number cases, role Artificial Intelligence (AI) imperative in current scenario. AI would be powerful tool fight against this predicting cases advance. Deep learning-based time series techniques are considered predict world-wide advance for short-term medium-term dependencies with adaptive learning. Initially, data pre-processing feature extraction made real world dataset. Subsequently, prediction cumulative confirmed, death recovered modelled Auto-Regressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM), Stacked (SLSTM) Prophet approaches. For long-term forecasting multivariate LSTM models employed. performance metrics computed results subjected comparative analysis identify most reliable model. From results, it evident that algorithm yields higher accuracy error less than 2% compared other algorithms studied metrics. Country-specific city-specific India Chennai, respectively, predicted analyzed detail. Also, statistical hypothesis correlation done on datasets including features like temperature, rainfall, population, total infected area population density during months May, June, July August find out best suitable Further, practical significance elucidated terms assessing characteristics, scenario planning, optimization supporting Sustainable Development Goals (SDGs).

Language: Английский

---

Deep learning via LSTM models for COVID-19 infection forecasting in India

PLoS ONE, Journal Year: 2022, Volume and Issue: 17(1), P. e0262708 - e0262708

Published: Jan. 28, 2022

The COVID-19 pandemic continues to have major impact health and medical infrastructure, economy, agriculture. Prominent computational mathematical models been unreliable due the complexity of spread infections. Moreover, lack data collection reporting makes modelling attempts difficult unreliable. Hence, we need re-look at situation with reliable sources innovative forecasting models. Deep learning such as recurrent neural networks are well suited for spatiotemporal sequences. In this paper, apply long short term memory (LSTM), bidirectional LSTM, encoder-decoder LSTM multi-step (short-term) infection forecasting. We select Indian states hotpots capture first (2020) second (2021) wave infections provide two months ahead forecast. Our model predicts that likelihood another in October November 2021 is low; however, authorities be vigilant given emerging variants virus. accuracy predictions motivate application method other countries regions. Nevertheless, challenges remain reliability difficulties capturing factors population density, logistics, social aspects culture lifestyle.

Language: Английский

---

A tree based eXtreme Gradient Boosting (XGBoost) machine learning model to forecast the annual rice production in Bangladesh

PLoS ONE, Journal Year: 2023, Volume and Issue: 18(3), P. e0283452 - e0283452

Published: March 27, 2023

In this study, we attempt to anticipate annual rice production in Bangladesh (1961–2020) using both the Autoregressive Integrated Moving Average (ARIMA) and eXtreme Gradient Boosting (XGBoost) methods compare their respective performances. On basis of lowest Corrected Akaike Information Criteria (AICc) values, a significant ARIMA (0, 1, 1) model with drift was chosen based on findings. The parameter value shows that positively trends upward. Thus, found be significant. other hand, XGBoost for time series data developed by changing tunning parameters frequently greatest result. four prominent error measures, such as mean absolute (MAE), percentage (MPE), root square (RMSE), (MAPE), were used assess predictive performance each model. We measures test set comparatively lower than those Comparatively, MAPE (5.38%) (7.23%), indicating performs better at predicting Bangladesh. Hence, Therefore, performance, study forecasted next 10 years According our predictions, will vary from 57,850,318 tons 2021 82,256,944 2030. forecast indicated amount produced annually increase come.

Language: Английский

---

Covid-19 Pandemic's Impact on Eating Habits in Saudi Arabia

Deleted Journal, Journal Year: 2020, Volume and Issue: 9(3)

Published: July 28, 2020

Background COVID-19 virus has been reported as a pandemic in March 2020 by the WHO. Having balanced and healthy diet routine can help boost immune system, which is essential fighting viruses. Public Health officials enforced lockdown for residents resulting dietary habits change to combat sudden changes. Design methods A cross-sectional study was conducted through an online survey describe impact of on eating habits, quality quantity food intake among adults Saudi Arabia. SPSS version 24 used analyze data. Comparison between general before during ordinal variables performed Wilcoxon Signed Rank test, while McNemar test nominal variables. The paired samples t-test compare total scores periods. Results 2706 residing Riyadh completed survey. majority (85.6%) respondents homecooked meals daily basis compared 35.6% (p<0.001). mean score slightly higher (p=0.002) period (16.46±2.84) (16.39±2.79). (p<0.001) (15.70±2.66) (14.62±2.71). Conclusion Dietary have changed significantly residents. Although some good increased, compromised. must focus increased awareness pandemics avoid negative consequences. Future research recommended better understand using detailed frequency questionnaire.

Language: Английский

---

Comparison of ARIMA, ETS, NNAR, TBATS and hybrid models to forecast the second wave of COVID-19 hospitalizations in Italy

The European Journal of Health Economics, Journal Year: 2021, Volume and Issue: 23(6), P. 917 - 940

Published: Aug. 4, 2021

The coronavirus disease (COVID-19) is a severe, ongoing, novel pandemic that emerged in Wuhan, China, December 2019. As of January 21, 2021, the virus had infected approximately 100 million people, causing over 2 deaths. This article analyzed several time series forecasting methods to predict spread COVID-19 during pandemic's second wave Italy (the period after October 13, 2020). autoregressive moving average (ARIMA) model, innovations state space models for exponential smoothing (ETS), neural network autoregression (NNAR) trigonometric model with Box-Cox transformation, ARMA errors, and trend seasonal components (TBATS), all their feasible hybrid combinations were employed forecast number patients hospitalized mild symptoms intensive care units (ICU). data February 2020-October 2020 extracted from website Italian Ministry Health ( www.salute.gov.it ). results showed (i) better at capturing linear, nonlinear, patterns, significantly outperforming respective single both series, (ii) numbers COVID-19-related hospitalizations ICU projected increase rapidly mid-November 2020. According estimations, necessary ordinary beds expected double 10 days triple 20 days. These predictions consistent observed trend, demonstrating may facilitate public health authorities' decision-making, especially short-term.

Language: Английский

---

Impact of lockdowns on the spread of COVID-19 in Saudi Arabia

Informatics in Medicine Unlocked, Journal Year: 2020, Volume and Issue: 20, P. 100420 - 100420

Published: Jan. 1, 2020

Epidemiological models have been used extensively to predict disease spread in large populations. Among these models, Susceptible Infectious Exposed Recovered (SEIR) is considered be a suitable model for COVID-19 predictions. However, SEIR its classical form unable quantify the impact of lockdowns. In this work, we introduce variable system equations study various degrees social distancing on disease. As case study, apply our modified initial data available (till April 9, 2020) Kingdom Saudi Arabia (KSA). Our analysis shows that with no lockdown around 2.1 million people might get infected during peak 2 months from date was first enforced KSA (March 25th). On other hand, Kingdom's current strategy partial lockdowns, predicted number infections can lowered 0.4 by September 2020. We further demonstrate stricter level curve effectively flattened KSA.

Language: Английский

---

Modeling and forecasting of COVID-19 using a hybrid dynamic model based on SEIRD with ARIMA corrections

Infectious Disease Modelling, Journal Year: 2020, Volume and Issue: 6, P. 98 - 111

Published: Dec. 3, 2020

The outbreak of novel coronavirus (COVID-19) attracted worldwide attention. It has posed a significant challenge for the global economies, especially healthcare sector. Even with robust system, countries were not prepared ramifications COVID-19. Several statistical, dynamic, and mathematical models COVID-19 including SEIR model have been developed to analyze infection its transmission dynamics. objective this research is use public data study properties associated pandemic develop dynamic hybrid based on SEIRD ascertainment rate automatically selected parameters. proposed consists two parts: modified ARIMA models. We fit parameters against historical values infected, recovered deceased population divided by rate, which, in turn, also parameter model. Residuals first recovered, populations are then corrected using can input real-time provide long- short-term forecasts confidence intervals. was tested validated US COVID statistics dataset from Tracking Project. For validation, we unseen recent statistical data. five common measures estimate prediction ability: MAE, MSE, MLSE, Normalized MSE. proved great ability make accurate predictions patients. output be used government, private sectors, policymakers reduce health economic risks significantly improved consumer credit scoring.

Language: Английский

---

Forecasting fully vaccinated people against COVID-19 and examining future vaccination rate for herd immunity in the US, Asia, Europe, Africa, South America, and the World

Applied Soft Computing, Journal Year: 2021, Volume and Issue: 111, P. 107708 - 107708

Published: July 14, 2021

Language: Английский

---

Application of machine learning in the prediction of COVID-19 daily new cases: A scoping review

Heliyon, Journal Year: 2021, Volume and Issue: 7(10), P. e08143 - e08143

Published: Oct. 1, 2021

COVID-19 has produced a global pandemic affecting all over of the world. Prediction rate spread and modeling its course have critical impact on both health system policy makers. Indeed, making depends judgments formed by prediction models to propose new strategies measure efficiency imposed policies. Based nonlinear complex nature this disorder difficulties in estimation virus transmission features using traditional epidemic models, artificial intelligence methods been applied for spread. importance machine deep learning approaches spreading trend, present study, we review studies which used these predict number cases COVID-19. Adaptive neuro-fuzzy inference system, long short-term memory, recurrent neural network multilayer perceptron are among mostly regard. We compared performance several Root means squared error (RMSE), mean absolute (MAE), R

Language: Английский

---