Table 7 AI/machine learning models for forecasting the COVID-19 epidemic.
From: Forecasting the spread of COVID-19 based on policy, vaccination, and Omicron data
Authors | Target Regions | Response variables | Predictor variables | Prediction models |
|---|---|---|---|---|
Pinter et al. (2020) | Single region | Daily new cases Mortality rate | Single variable (past observation) | Multiple models (MLP-ICA, ANFIS) |
Saba and Elsheikh (2020) | Single region | Cumulative cases | Single variable (past observation) | Multiple models (ARIMA, NARANN) |
Ramazi et al. (2021) | Single region | Daily new cases Daily new deaths | Multiple variables (daily COVID-19 tests daily temperature daily precipitation Google mobility) | Multiple models (LaFoPaFo,UCLA-SuEIR, STH-3PU, etc.) |
Gomez-Cravioto et al. (2021) | Single region | Daily new cases Daily new deaths | Multiple variables (weather information—temperature, UV index, humidity, etc Google mobility) | Multiple models (logistic growth curve, ARIMA, LSTM, etc.) |
Elsheikh et al. (2021) | Multiple regions | Cumulative cases Cumulative recovered Cumulative deaths | Single variable (past observation) | Multiple models (ARIMA, NARANN, LSTM) |
Al-qaness et al. (2021) | Multiple regions | Daily new cases | Single variable (past observation) | Multiple models (ANFIS, ANFIS-MPA, ANFIS-CMPA) |
Meakin et al. (2022) | Single region | Daily hospitalization | Single variable (Daily new cases with time lag) | Multiple models (baseline, Timeseries ensemble, ARIMA, etc.) |
