Fig. 3
From: Identification of effective spreaders in contact networks using dynamical influence
Results from SI simulations (\(\tau = 1\)) of real-world contact networks using four selected spreaders. The contact networks are, a a hospital ward (Vanhems et al. 2013), b, c, a primary school (Stehlé et al. 2011), d a high school (Salathé et al. 2010), e, f a workplace (Génois et al. 2015; Génois and Barrat 2018) and, g a conference (Génois and Barrat 2018). Times are recorded from 100 SI simulations when 25%, 50%, 75%, and 100% of network are infected and then normalised with the mean of all times recorded at each percentage. From left to right; times are detailed for the weighted degree (no neighbours) selection, then the eigenvector-based selections for initial rate (Initial), average performance (Averaged), and time to all infected (End). The y-axes of a, c, f, g are cropped to exclude extreme values