Evaluation of Cybersecurity Performance in Intrusion Detection Systems Based on Fuzzy VIKOR and the Nash Equilibrium

Intrusion detection system (IDS) is a kind of network security device that can monitor network traffic immediately and send alarm or take active response measures when finding suspicious transmission. As malicious attacks continue to change and occur in very large numbers, it is essential to evaluate the cybersecurity performance and select a flexible and effective IDS matching the corresponding protection requirements. In this study, the evaluation of cybersecurity performance in IDS is established as a fuzzy multi-attribute decision making (MADM) problem, and the performance of IDS is comprehensively evaluated in 8 indexes, such as spam detection, phishing identification, anomaly detection, etc. In MADM, fuzzy analytic hierarchy process is used to reflect the relative importance of each evaluation index based on the actual requirements of protection and the preference of domain experts, i.e. the subjective attribute weights. To avoid subjectivity randomness of preferences expression, the fuzzy entropy weight method based on α-cut is used to measure intrinsic information in decision making matrix and determine the objective weight of each evaluation index. And on this basis, the Nash Equilibrium is introduced to integrate the subjective and objective attribute weights, so as to retain more raw information about the different types of attribute weights. Finally, the fuzzy VIKOR is used to comprehensively evaluate each IDS and select the optimal IDS. Experimental results verify that the proposed method can be used to choose the best IDS flexibly and efficiently.