Robustness Analysis of Social Network based on a Dynamic Model

Due to the high vulnerability of today’s social systems against disruptions, it has become critical to measure and manage the robustness of social networks. In this study the relationship between social network’s topology and its robustness in the presence of cascading failures is examined. In the attempt to help practice in increasing the robustness of social systems, the concept of the robustness of social systems is discussed and defined first. And then we propose a new dynamic model of social systems to investigate the robustness of social systems. We theorize that network characteristics such as average path length, node degree, and size of the largest connected component in social networks and so on. Based on the model, a good evaluation method is proposed to investigate the robustness of social networks against the cascading failures of component nodes. Finally, the damage degree of social networks considering cascading failures is investigated through our model. The simulation results confirm that there is a positive relationship between the robustness of social systems and social system performance, while a negative relationship between the topology structure and the robustness of social systems. By the dynamic model, we not only identify the potential key nodes, but also obtain the parameter value which can make the social network reach strongest robustness level. Our work develops a new methodology to the real world application. This research is helpful to increase the robustness of such complex systems and tackle the optimization problem of them.