Article

Empirical Evaluation of Cyber Defense Mechanism Effectiveness Using Machine Learning for Strengthening Digital Security Regulations 

---

Abstract

This study presents an empirical framework for evaluating the effectiveness of cyber defense mechanisms using machine learning to support the development of evidence-based digital security regulation. The research utilizes a multi-year dataset on global cybersecurity incidents from 2015 to 2024, incorporating both technical and contextual variables such as attack type, target industry, defense mechanism, financial loss, and resolution time. By applying advanced supervised learning algorithms, including ensemble-based models, the study achieved an overall accuracy of 83.7 percent and a macro-averaged F1-score of 0.837. The model demonstrated strong performance in classifying high- and low-impact incidents while maintaining acceptable precision for medium-impact cases. Feature importance analysis identified financial loss, loss per user, and resolution efficiency as the most influential factors affecting defense effectiveness. The findings indicate that machine learning can provide a transparent, quantitative approach to measuring cybersecurity performance, bridging the gap between technical evaluation and legal compliance. From a regulatory perspective, the results suggest that data-driven models can inform the formulation of standardized benchmarks for digital security compliance and accountability. This research contributes to the intersection of technology and law by offering a methodological foundation for integrating predictive analytics into cyber law enforcement and international cybersecurity governance.

Keywords: Machine Learning, Cyber Defense, Digital Security Regulation, Cyber Law, Predictive Analysis

How to Cite:

Prakash, S., (2026) “Empirical Evaluation of Cyber Defense Mechanism Effectiveness Using Machine Learning for Strengthening Digital Security Regulations ”, Journal of Cyber Law 2(1). doi: https://doi.org//JCL.154