Analysis and Forecast of the COVID-19 Spreading Curve for the Resumption of In-person Classes

With the COVID-19 pandemic, maintaining social distancing is particularly important in daily life. In recently, indoor situations such as face-to-face teaching for university restart are tried to make feasible suggestions depend on the spread of the COVID-19. In this research, we analyze and forecast the COVID-19 spreading curve of the resumption of in-person classes at university by the graph structure with the spread weight of edges based on each student’s relation. Our approach is based on the effectiveness of three distancing strategies designed to keep the curve flat and aid make the spread of the COVID-19 controllable. By detecting the possibility of student relation by three strategies, we can analyze the COVID-19 spreading curve by Graph Neural Network(GNN) and SIR model. The SIR model is a simple model that considers a population that belongs to one of the following states: Susceptible (S), Infected (I), and Recovered (R), and we calculate the contagion rate of the pathogen. In this article, we discuss two types of Open Group and Closed Group on university campuses and analyze face-to-face lectures, indoor social activities, and campus cafeterias. To verify the effectiveness of our two types of group, we simulated with the random infection curve by graph neural network model. The simulation analysis results show that our social distancing strategies can reduce the risk of COVID-19 transmission after school restarts.