Identifying influential spreaders in complex networks

 

MAKSIM KITSAK

UC SAN DIEGO

 

Networks portray a multitude of interactions through which people meet, ideas are spread, and infectious diseases propagate within a society. The complexity of such systems implies that innovative ideas and important news can reach remote areas with surprising speed. Identifying the most efficient ``spreaders'' through a reliable measure based on network theory is an important step to optimize the use of available resources and ensure the more efficient spread of information. Here we show that, in contrast to common belief, in many circumstances the most influential spreaders in a social network do not correspond to the best connected people (system hubs) or to the most central people (top betweenness centrality). Instead, a network analysis shows: (i) The people that affect the largest portion of the population in a spreading process starting from a single spreader are located within the core of the network identified by a $k$-shell decomposition analysis. (ii) When multiple spreading origins are considered simultaneously, the distance between them becomes the crucial parameter that determines the extend of the spreading. Furthermore, we find that--- in the case of infections that do not confer immunity on recovered individuals--- the infection persists in the high $k$-shell layers of the network under conditions where hubs may not be able to preserve the infection. The present network analysis helps to identify the most influential people, providing a plausible route for an optimal design of efficient dissemination strategies.