Self-Adaptive AI-Driven Intelligent Multimodal Anomaly Detection Framework for Enhancing Cybersecurity in Cloud-Based. NET Applications

Abstract

Accurate and flexible anomaly detection is crucial for preserving security and service availability since the quick uptake of cloud-based.NET apps have increased exposure to sophisticated cyber threats. To improve cybersecurity in cloud-based.NET systems, this article suggests a Self-Adaptive AI-Driven Intelligent Multimodal Anomaly Detection Framework. To provide thorough threat detection, the framework combines a variety of data sources, such as network traffic, application logs, user activity, and system performance indicators. To extract spatial-temporal behavioral patterns and detect sophisticated cyberattack operations, a hybrid anomaly detection approach that combines entropy-based traffic analysis, graph-based dependency modeling, and rule-driven event correlation is utilized. By continuously updating detection parameters in response to changing threat conditions, the self-adaptive module increases resilience to dynamic and zero-day threats. When compared to traditional anomaly detection methods, experimental evaluation shows better detection accuracy, lower false-positive rates, and quicker response times. For sophisticated cybersecurity protection in contemporary cloud-based.NET applications, the suggested framework offers a practical and scalable solution.