Role of Individual Activity in Rumor Spreading in Scale-free Networks

Extensive empirical results show that a large variety of social networks, for instance, Facebook, Wechat, and QQ, exhibit a power-law degree distribution, which are typically referred to as scale-free networks. Previous studies mainly concern the impact of spreading and annihilation rate of a rumor on its spreading process in the undirected scale-free networks. In this paper, we investigate the impact from individual activity. The individual activity varies widely in social networks, which determines how likely a spreader would disseminate a received rumor out to its neighbors. For the Barabá si-Albert scale-free networks, our analytical and simulation results surprisingly reveal that the final informed scale of a rumor is unrelated to the individual activity. To provide a timely estimation of the final informed scale, we propose an accelerated Monte-Carlo algorithm, which not only drastically promotes the efficiency of Monte-Carlo simulation, but also guarantees its validity. Our work is thus helpful to further understand the information dissemination on real-world social networks. The accelerated Monte-Carlo algorithm is also capable of further promoting related studies on epidemic-style dynamical processes.