Intelligent dynamic cybersecurity risk management framework with explainability and interpretability of AI models for enhancing security and resilience of digital infrastructure
Computational study demonstrates improved vulnerability prioritization in digital infrastructure data, indicating enhanced dynamic cybersecurity risk assessment.
Key Points
To develop and evaluate a dynamic cybersecurity risk management framework that integrates changing infrastructure parameters and explainable artificial intelligence to prioritize software vulnerabilities.
Formulated the dynamic cybersecurity risk management (d-CSRM) framework, incorporating dynamic factors such as vulnerability exploitation and asset dependencies.
Engineered a hybrid AI model combining linear regression and deep learning with integrated model explainability and interpretability tools.
Tested and validated the vulnerability prioritization framework against the CVEjoin benchmark dataset.
The hybrid model successfully identified and prioritized the most critical system vulnerabilities.
Model interpretability mechanisms effectively extracted key risk features—including exploit type, exploit platform, and impact—to justify decision-making.