Time dependent correlations between the probability of a node being infected and its centrality measures

Pandemics are a growing world-wide threat for all societies. Throughout history, various infectious diseases presented widely spread damage to human life, economic viability and general well-being. The scale of destruction of the most recent pandemic, COVID-19, has yet to be seen. This work aims to introduce intervention methodology for the prevention of global scale spread of infectious diseases. The proposed method combines time-dependent infection spreading data with the social connectivity structure of the society. SIR model simulations provided the dynamic of contamination spread in different sets of network data. Seven centrality measures parameterized the local and global importance of each node in the underlying network. At each time step the calculated values of the correlations between node infection probability and node centrality values are analyzed. Calculations show that correlations increase at the beginning of infection spread and reaches its highest value when spreading starts to become an epidemic. The peak is at the very early stages of the spreading; and with this analysis, it is possible to predict the node infection probability from time-dependent correlations data.