Controlling contagion processes in activity-driven networks

S. Liu, N. Perra, M. Karsai, A. Vespignani
Physical Review Letter, 112:118702, (2014)
March 19, 2014

Abstract

The vast majority  of strategies aimed at controlling contagion processes on networks consider  the connectivity pattern of the system either quenched or annealed. However,  in the real world, many networks are highly dynamical and evolve, in time,  concurrently with the contagion process. Here, we derive an analytical  framework for the study of control strategies specifically devised for a  class of time-varying networks, namely activity-driven networks. We develop a  block variable mean-field approach that allows the derivation of the  equations describing the coevolution of the contagion process and the network  dynamic. We derive the critical immunization threshold and assess the  effectiveness of three different control strategies. Finally, we validate the  theoretical picture by simulating numerically the spreading process and  control strategies in both synthetic networks and a large-scale, real-world,  mobile telephone call data set.