Estimation of Epidemic State using GNNs
Utilization of GCNs to predict the state (S/I/R) of a person in a small-world network during an epidemic
Abstract: Epidemic forecasting provides an opportunity to predict geographic disease spread as well as case counts to better inform public health interventions when outbreaks occur. In this project, we firstly explore the work already done in the domain of epidemics, using both classical machine learning and deep learning. Further, we demonstrate the spread of epidemics using the famous SIR model on random graphs generated by the Erdős–Rényi-Gilbert, Watts-Strogatz and Barabási–Albert models. Finally, we construct a graph neural network followed by a fully connected network to predict the category (safe/infected/recovered) a person might be in during an epidemic.