Analytics and measuring the vulnerability of communities for complex network security

Analytics and measuring the vulnerability of communities for complex network security

연구 분야: Infrastructure

학회: International Journal of Data Science and Analytics

Complex networks are used as a tool for data representation and model a wide range of natural and artificial systems and networks with their many interdependent components. The security of complex networks is one of the key issues in network analysis to prevent accidental or intentional attacks. Community structure is an inherent and natural characteristic of complex networks and reflects their potential performance. This feature increases resilience and reduces the propagation of security attacks and cascading failures in a network. Since vulnerable network structures play an important role in the increase of security threats, in this paper, the vulnerability of each community is measured based on the topological characteristics, and the propagation process of security attacks is simulated in each network community. The proposed security strategy is such that first the community structures of the network are identified. The vulnerable communities are then evaluated for appropriate configuration using the presented vulnerability criterion, which is based on intra-community and inter-community connections. Finally, the propagation and expansion of security threats in the communities are simulated and analyzed according to their vulnerability using the SIRS (susceptible–infected–recovered–susceptible) model and the calculation of the basic reproduction number. The experiments are examined on six real-world complex networks. The simulation results show that the proposed security strategy is effective in the role and identification of vulnerable structures to improve complex network security.