| Autor: |
Kartikasari, Puspita, Yasin, Hasbi, Maruddani, Di Asih I. |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2023, Vol. 2683 Issue 1, p1-6, 6p |
| Abstrakt: |
Predictive model of the spread of COVID-19 play an important role in epidemiological study and government effort to deal with this pandemic. However, some COVID-19 prediction models are dominated by constant model parameters so they cannot reflect the actual situation of the spread of COVID-19. This study presents a method for dynamic prediction of the spread of COVID-19 by considering time-dependent model parameters. Neural Network is a computer technique that produce predictive models with a simpler and more flexible form. This is because the resulting model from this method is shaped by many processing units and the flexibility comes from iterating between units. One of the applications of Neural Networks is for time series data prediction algorithms. In this study, the Neural Network was applied to predict the number of deaths caused by the COVID-19 pandemic. From the results obtained, it can be concluded that the prediction model for cases of death due to COVID-19 has been successfully created in its entirety with the best architecture obtained is (12, 10, 3) from the test set between testing and trending data. In addition, the architectural model has a Mean Squared Error (MSE) of 0.286. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
