Integrated Neutrosophic methodology and Machine Learning Models for Cybersecurity Risk Assessment: An exploratory study

Abstract

Information technology security, or Cybersecurity, guards against hostile cyberattacks on computers, mobile devices, servers, electronic systems, and networks. Cybersecurity risks have been a significant concern for any vital digital infrastructure in recent years, and different online cyberattacks are also becoming a significant problem for society. Consequently, it's critical to adopt technology created to provide cybersecurity. However, one should consider the associated hazards while selecting among Cybersecurity systems. We have developed a multi-criteria decision-making (MCDM) approach based on a single-valued neutrosophic set (SVNS). This allows specialists more latitude in assessing the criteria and alternatives using language and overcoming uncertain information. The VIKOR is an MCDM methodology used to rank the other options. The VIKOR method is integrated with the neutrosophic set. There are 18 criteria, and 10 alternatives are used in this study. The sensitivity analysis and comparative analysis are conducted in this study. The sensitivity analysis results show the alternatives' rank is stable under different cases. The comparative analysis compares the suggested method with other MCDM methods. The comparative analysis shows the suggested method was effective compared with other MCDM methods. Machine learning methods predict the type of attack in Cybersecurity. This study uses Three machine learning methods: decision tree, random forest, and support vector machine.