Effect of observation time on source identification of diffusion in complex networks

Abstract

This paper examines the effect of the observation time on source identification of a discrete-time susceptible-infected-recovered diffusion process in a network with snapshot of partial nodes. We formulate the source identification problem as a maximum likelihood (ML) estimator and develop a statistical inference method based on Monte Carlo simulation (MCS) to estimate the source location and the initial time of diffusion. Experimental results in synthetic networks and real-world networks demonstrate evident impact of the observation time as well as the fraction of the observers on the concerned problem.