Container Security Intelligence: Leveraging Machine Learning for Anomaly Detection in Containerized Applications

Abstract

This research explores the application of Machine Learning (ML) techniques for anomaly detection in containerized applications. The proposed approach utilizes by applying machine learning techniques for anomaly detection in containerized applications. The proposed methodology integrates container runtime metrics, system call analysis, and network traffic patterns to build a predictive model capable of identifying abnormal behavior within containers. To evaluate the effectiveness of this approach, extensive experiments were conducted using real-world containerized applications and datasets. The experiment focused on assessing the accuracy of anomaly detection and the computational efficiency of the proposed model. The results demonstrated that this approach achieved a remarkable accuracy rate of 95% in detecting anomalies, while maintaining high computational efficiency, with minimal overhead on the container runtime environment. This research contributes valuable insights into bolstering the security of containerized applications, offering practical solutions for real-world deployment scenarios. The research also evaluates the accuracy and efficiency of the proposed approach and discusses its potential impact on enhancing container security in real-world scenarios.