The selection of critical set for network attack targets: from the perspective of attacker

With the development of networks, topics related to network security are receiving increasing attention from scholars. Graph-based parameters are defined to measure the vulnerability of a network from a topological perspective, and the calculation of these parameters often requires finding the corresponding cut set $$S\subset V(G)$$. From the attacker’s perspective, it is beneficial to achieve the maximum destructive effect using the minimum cost. The cut set S corresponding to the vulnerability parameter is the optimal target site chosen by the attacker, and such a cut set is called a critical set. To retrieve the critical set of general graphs, this article modifies the traditional I-Ching divination evolution algorithm, proposes the circle representation of individuals, and obtains the crossover operator of hexagrams. In addition, the local versions of graph-based parameters are introduced for application backgrounds with specific attack targets. Finally, the effectiveness of the proposed algorithm was verified through experiments.