KRONOS: Knowledge-driven Recursive Operations for Network Operations Security - A Self-Morphing AI Framework for Adaptive Cybersecurity

作者：Swayam Singh, Akhil Chauhan, A. Rawat · 发表于：2026 IEEE 1st International Conference on Intelligent Technologies for a Sustainable and Inclusive Future (ICITSIF) · 年份：2026 · DOI：10.1109/icitsif69060.2026.11608368

The speed at which contemporary cybersecurity threats are changing necessitates the use of adaptive defenses that can quickly adjust to new attack trends. This paper presents KRONOS (Knowledge-driven Recursive Operations for Network Operations Security), a self-morphing AI framework composed of three interconnected modules: ORDER (defense), CHAOS (attack simulation), and BALANCE (evolutionary adaptation). To identify potential threats in network traffic, the ORDER module applies anomaly detection methods such as Isolation Forest and related machine learning classifiers. The BALANCE module is designed to support genetic algorithms and reinforcement learning for automated tuning, while the CHAOS module generates simulated attack traffic to continuously stress-test the defense mechanisms. Our implementation demonstrates strong detection performance on the KDD Cup 1999 dataset, achieving high accuracy and low false positive rates in preliminary experiments.