EGAO-ICONIDS: Adaptive Container-Aware Cybersecurity Monitoring System for Kubernetes Using Enhanced Feature Selection and Conformer Networks
Published: 18 September 2026| Version 1 | DOI: 10.17632/n7hgxcxx2m.1
Contributors:
Dharani Modugumudi, Description
In conclusion, employing the suggested EGAO-ICONIDS to identify and monitor threats in Kubernetes is a very efficient method of enhancing the security of containerized applications. Attacks are becoming more likely as Kubernetes becomes increasingly popular for managing and deploying containerized workloads. EGAO-ICONIDS offers a potent method for tracking and evaluating network traffic, enabling extremely detailed and instantaneous insight into the system's behavior. This allows us to identify attacks and notify the system.
Files
Categories
Deep Learning