14 August 2024 in Applied Soft Computing
1 December 2024 in Chaos: An Interdisciplinary Journal of Nonlinear Science
Critical infrastructure systems (CIS) are closely related to human life. Attacks against CIS occur frequently, making accurate and effective protection of CIS essential. Vulnerability assessment is the primary issue to achieving this goal. The interconnected characteristic of CIS means that it is best represented by a multi-layer network, but a uniform model is absent. Game theory offers a suitable framework for researching intelligent confrontation. Previous research combining game theory and network science mainly focuses on a single-layer network and lacks a comprehensive assessment that combines qualitative and quantitative aspects of vulnerability. In this paper, we apply the Stackelberg game to the multi-layer network and comprehensively assess vulnerability based on game equilibrium to realize accurate protection. We first present a method for constructing a general model of the multi-layer network and introduce a multi-layer weighted factor to extend topological attribute metrics. Then, we design a Stackelberg game model for the multi-layer network. Furthermore, we qualitatively analyze the impact of multi-layer network characteristics on vulnerability and propose a method to quantify vulnerability. Experiments show that the vulnerability of the multi-layer network is greatly influenced by the multi-layer weighted factor, single-layer network type, and inter-layer coupling method. The quantitative value of network vulnerability does not rely entirely on the topology attributes on which the attack and defense costs depend but also relates to the resources available. Our work provides an adaptive model for CIS and gives a new approach to developing accurate protection based on comprehensive vulnerability assessment, which deserves further study.
1 January 2024 in Chinese Journal of Electronics
Link prediction utilizes accessible network information to complement or predict the network links. Similarity is an important prerequisite for link prediction which means links more likely occurs between two similar nodes. Existing methods utilize the similarity of nodes but neglect of network structure. However the link direction leads to a far more complex structure and contains more information useful than the undirected networks. Most classic methods are difficult to depict the distribution of the network structure with incidental direction so the similarity characteristics of the network structure itself are lost. In this respect, a new method of local structure entropy is proposed to depict the directed structural distribution characteristics, which can be used to evaluate the degree of local structural similarity of nodes and then applied to link prediction methods. Experimental results on 8 real directed networks show that this method is effective for both area under the receiver operating characteristic curve (AUC) and ranking-score measures, and improved predictive capacity of the baseline methodology.