Abstract

The increasing complexity and volume of cyber threats necessitate intelligent and adaptive security solutions for modern cloud infrastructures. This research explores the integration of Machine Learning (ML) techniques with Microsoft Azure Security Center (ASC) to enhance threat detection, risk mitigation, and proactive security management. Azure Security Center, a unified infrastructure security management system, offers built-in ML-driven analytics for anomaly detection, behavioral analysis, and automated threat response. The study investigates how ML algorithms, such as anomaly detection models, decision trees, and neural networks, are utilized within ASC to detect potential threats across hybrid and multi-cloud environments. By analyzing telemetry data, network behavior, and resource configurations, ASC’s ML capabilities help in identifying patterns indicative of malicious activity, zero-day exploits, and insider threats in near real-time. The research further evaluates the effectiveness, scalability, and accuracy of ASC’s MLdriven threat detection compared to traditional rule-based systems. Case studies and simulated attack scenarios are used to demonstrate ASC’s predictive capabilities and response time improvements. The findings highlight the value of embedding ML into cloud-native security platforms for achieving faster threat detection, reducing false positives, and enabling a more resilient security posture. This study contributes to the growing field of intelligent cloud security by offering insights into the practical deployment of machine learning within enterprise-grade security ecosystems like Azure.

Keywords: Anomaly detection, Network behavior, Neural networks, Predictive analytics, Threat intelligence