A similarity-based M(1,2,3) approach for rating the security of network systems

Abstract

It is a multiple criteria problem to evaluate the security of information networks. However, one key issue which is not fully considered in extant studies is how to obtain the values of these multiple evaluation criteria, and in most evaluation models some redundant criteria values are calculated. Motivated by this, in the work we first use the pairwise comparison method to attain the values of each evaluation aspect of the security of information networks, and then develop a novel similarity-based M(1, 2, 3) approach to rate the security of alternative networks. Application study shows that the proposed approach could rate the overall security of alternative information networks and recognize the most vulnerable aspects of each information network, which provides good supports for making policies to enhance the security of alternative information networks.