Detection of early-stage infection spreading in complex social networks: a simulation study

Abstract

With the current COVID-19 situation, simulation of the early-stage infection spreading has become more and more important in the epidemic prevention and control. With such simulations, the government and public health department could be able to track infectious cases and predict the scale of epidemic. It is also important when deciding how much public expenditure to be spent on epidemic precautions.

In this project, the author will build up a scale-free network model by Python, and conduct infection simulation under the existence of hidden link. With different hidden link rate during the contact tracing progress, the author could evaluate the deviations brought about by the hidden link and other key factors which could affect the accuracy of contact tracing. As a result of hidden links, to control the early-stage pandemic, the health authority would need to quarantine more close contacts. Within the same network model, the author will find out by how many percent the quarantine number should be included additionally. To simulate the spreading of infectious disease, research on epidemiology was also carried out. The choice of key parameters will also be covered.

With the help of the network model, the author will also carry out a few simulations on factors which will affect virus spreading, such as the existence of super spreader, the necessity of wearing a mask and the necessity of maintaining social distance in public places. The purpose of this part is to discuss the effectiveness of several anti-epidemic measures at this stage and offer firm evidence with specific statistics.

In the end, the author will list out several limitations of the network model and discuss several possible solutions and improvements that could be made to the model.