Networks are used to represent complex systems in the real world. Recently, the focus of interest in the area of network science has shifted to the controllability of complex networks. In this context, the new concept of a subset of nodes called driver nodes is becoming pronounced in the network world. Driver nodes belong to the intersection of network science and the control theory of engineering. The growing interest in this field has resulted in an opportunity for scientists to explain how to control the dynamics of complex systems. The scope of this study is to directly test (1) the fractions of driver nodes distributions of real networks and fully randomized networks and (2) the statistically significant difference in the mean fractions of driver nodes between natural and manmade networks and plus between natural and fully randomized networks. On the basis of the sample results, it is found that whereas real networks follow a largest extreme value distribution, fully randomized networks follow a gamma distribution. In addition, whereas a statistically significant difference was found between natural and manmade networks, no difference was found between natural and fully randomized networks.
Keywords: Networks, Complex Systems, Driver Nodes, Controllability, Critical Nodes.