LSTM-based Forecasting using Policy Stringency and Time-varying Parameters of the SIR Model for COVID-19
19th IEEE International Colloquium on Signal Processing and Its Applications, CSPA 2023
; : 111-116, 2023.
Artículo
en Inglés
| Scopus | ID: covidwho-2316923
ABSTRACT
Accurate forecasting of the number of infections is an important task that can allow health care decision makers to allocate medical resources efficiently during a pandemic. Two approaches have been combined, a stochastic model by Vega et al. for modelling infectious disease and Long Short-Term Memory using COVID-19 data and government's policies. In the proposed model, LSTM functions as a nonlinear adaptive filter to modify the outputs of the SIR model for more accurate forecasts one to four weeks in the future. Our model outperforms most models among the CDC models using the United States data. We also applied the model on the Canadian data from two provinces, Saskatchewan and Ontario where it performs with a low mean absolute percentage error. © 2023 IEEE.
COVID-19; epidemiological model; Long Short-Term Memory; policy pruning; Adaptive filtering; Adaptive filters; Brain; Decision making; Forecasting; Stochastic models; Stochastic systems; Decision makers; Epidemiological modeling; Health-care decisions; Infectious disease; Nonlinear adaptive filter; Saskatchewan; SIR model; Stochastic-modeling; Time varying parameter
Texto completo:
Disponible
Colección:
Bases de datos de organismos internacionales
Base de datos:
Scopus
Idioma:
Inglés
Revista:
19th IEEE International Colloquium on Signal Processing and Its Applications, CSPA 2023
Año:
2023
Tipo del documento:
Artículo
Similares
MEDLINE
...
LILACS
LIS