In 2020, COVID-19 became a global pandemic and has been negatively impacting the world ever since. One of the ways to help control the spread of disease is to forecast its growth. Forecasting the spread of disease, though a daunting task, is necessary when a pandemic happens as it could help create social policies that could potentially mitigate the effects of the disease. This research aims to provide two disease prediction methods: the R0-based method and the gradient-descent method. The R0-based method produces short-term predictions by simulating the mobility of residents and utilizing the R0 coefficient. The gradientdescent method maps the linear regression model onto non-linear model in order to fit the exponential growth of the disease. Experimental results show that the R0-based method is accurate at forecasting during pandemic outbreaks. The gradient-descent method is able to study the spread of epidemics on a city-to-city scale through transport network, but with less accuracy than the R0-based technique.
Nguyen, Dinh Song An, "Two Methods of Forecasting the Spread of Disease Using R0 Coefficient and Gradient-Descent" (2022). CSB/SJU Distinguished Thesis. 27.