Automated Cyber Risk Quantification Using Multi-Modal AI Models
DOI:
https://doi.org/10.15680/IJCTECE.2024.0703012Keywords:
Cyber Risk Quantification, Multi-Modal AI Models, Machine Learning, Deep Learning, Anomaly Detection, Insider Threat Detection, Malware Detection, Risk Scoring, Real-Time Risk Assessment, Network Behavior, System Logs, Threat Intelligence, Copula Models, Data Breaches, Cybersecurity, Risk ExposureAbstract
Cyber risk quantification, or automation, has been used in making cybersecurity assessments who transform qualitative measures into quantifiable measures, which are then used to motivate decision-making, underwrite insurance, and operate risk prioritization in real time. Multi-modal AI models make use of various data sources, such as network behavior, system logs and anomaly scores to provide estimates of overall risk exposure. The current report discusses the way in which machine learning and deep learning methods can be utilized to support CRQ, provides examples of simulated and real-life situations, and assesses the performance of models using numerical data. The results indicate that the AI-driven CRQ can greatly enhance the capability of recognizing rather intricate threat cases and keep pace with the changing network dynamics, yet the difficulties are present with the quality of data and their interpretability in the models.
References
[1] A. Milenkoski, M. Vieira, S. Kounev, A. Avritzer & B. D. Payne, Evaluating Computer Intrusion Detection Systems: A Survey of Common Practices, ACM Comput. Surv., vol. 48, no. 1, Art. 12, 2015, doi: 10.1145/2808691.
[2] A. L. Buczak & E. Guven, A Survey of Data Mining and Machine Learning Methods for Cyber Security Intrusion Detection, IEEE Commun. Surveys & Tutorials, vol. 18, pp. 1153–1176, 2016, doi: 10.1109/COMST.2015.2494502.
[3] C. Yin, Y. Zhu, J. Fei & X. He, A Deep Learning Approach for Intrusion Detection Using Recurrent Neural Networks, IEEE Access, vol. 5, pp. 21954–21961, 2017, doi: 10.1109/ACCESS.2017.2762418.
[4] T. Aldwairi, D. Perera & M. A. Novotny, An Evaluation of the Performance of Restricted Boltzmann Machines as a Model for Anomaly Network Intrusion Detection, Computer Networks, vol. 144, Oct. 2018, doi: 10.1016/j.comnet.2018.07.025.
[5] “Machine Learning and Deep Learning Methods for Cyber Security Intrusion Detection,” Applied Sciences, vol. 9, no. 20, pp. 4396, 2019, doi: 10.3390/app9204396.
[6] M. Eling & K. Jung, Copula Approaches for Modeling Cross-Sectional Dependence of Data Breach Losses, Insurance: Mathematics and Economics, vol. 82, pp. 167–180, 2018, doi: 10.1016/j.insmatheco.2018.07.003.
