Investigation of recurrent networks LSTM in the problem of Covid-19 forecasting

Zaychenko, Yuriy; Zaichenko, Helen; Hamidov, Galib

doi:10.1109/csit52700.2021.9648696

Cited by 4 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gated recurrent units (GRU) is a typical RNN model [29][30][31]; the GRU network model and its structure is shown in Figure 1. Please rephrase the part for clarity, the structure of GRU is similar to the LSTM, but much simpler than the LSTM, which only contains two kinds of gates: the reset gate and the update gate.…”

Section: Methodsmentioning

confidence: 99%

Regional Economic Forecasting Method Based on Recurrent Neural Network

Liu

Zhu

Liu

et al. 2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Macroeconomic situation is the overall performance of the economic situation of a country and region. Making accurate forecasts of macroeconomic trends is of great significance for analyzing the success or failure of macroeconomic control policies, evaluating the quality of economic system operation, and correctly formulating future development planning strategies. The macroeconomic system is a nonlinear system, the environment is constantly changing, and additional disturbing factors directly affect the operation of the macroeconomic system, which has a great impact on the forecast results. The historical information required for macroeconomic modeling is unstable, unclear, and incomplete, which makes it very difficult to solve such problems with traditional forecasting methods. In response to the multivariate and nonlinear characteristics of macroeconomic forecasting, this paper proposes the application of artificial neural networks for forecasting. This paper introduces the recurrent neural network into the field of economic forecasting to solve the problems of the traditional BP (back propagation) neural network method. The experimental data are verified and the experimental results prove that the studied scheme based PSO-GRU improve the performance of economic forecasting.

show abstract

Section: Methodsmentioning

confidence: 99%

Regional Economic Forecasting Method Based on Recurrent Neural Network

Liu

Zhu

Liu

et al. 2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…Investigations of recurrent LSTM are carried out for medium-term forecasting of covid-19 pandemic indicators at the example of Ukraine [19]. The models of recurrent LSTM of two architectures (basic and extended) on different data sets are explored, after which the best models are found and corresponding forecast results are obtained.…”

Section: Related Workmentioning

confidence: 99%

Predictive Information Workflow of Forecasting Number of COVID-19 Confirmed Cases

Gam

Cao

Seow

2022

2022 IEEE 21st International Conference on Ubiquitous Computing and Communications (IUCC/CIT/DSCI/SmartCNS)

View full text Add to dashboard Cite

There are questions about how to accurately prepare with the correct number of resources for distribution in order to properly manage the healthcare resources (e.g., healthcare workers, Masks, ART-19 TestKit) required to tighten the grip on the COVID-19 pandemic. Mathematical and computational forecasting models have well served the means to address these questions, as well as the resulting advisories to governments. A workflow is proposed in this research, aiming to develop a forecasting simulation that makes accurate predictions on COVID-19 confirmed cases in Singapore. According to the analysis of the prior works, six candidate forecasting models are evaluated and compared in the workflow: polynomial regression, linear regression, SVM, Prophet, Holt's linear, and LSTM models. The study's goal is to determine the most suitable forecasting model for COVID-19 cases in Singapore. Two algorithms are also proposed to better compute the performance of two models: the order algorithm to determine optimal degree order for the polynomial regression model , and the optimizing algorithm for the Holt's linear model to calculate the optimal smoothing parameters. Observed from the experiment results with the COVID-19 dataset, the Prophet method model achieves the best performance with the lowest Root Mean Square Error (RMSE) score of 1557.744836 and Mean Absolute Percentage Error (MAPE) score of 0.468827, compared to the other five models. The Prophet method model achieving average accuracy range of 90% when forecasting the number of confirmed COVID-19 cases in Singapore for the next 87 days ahead. is chosen and recommended to be used as a system model for forecast the COVID-19 confirm cases in Singapore. The developed workflow will greatly assist the authorities in taking timely actions and making decisions to contain the COVID-19 pandemic.

show abstract