Abstract

With the applications like MalGAN, such cyberattacks enhanced with artificial intelligence (AI) in a broad way across cyber-defense lifecycles successfully take the vulnerabilities of systems at advantage, which are many as these are evading defenses nowadays. Therefore, this methodology proposed a new method which presents the approach of data fingerprinting and visualization for AI-Enhanced Cyber-Defense Systems (AIECDS) for efficiency in detection. AIECDS approach is built combining dynamic reinforcement learning, feature extraction and visualization with Hilbert curves and tornado graphs, real-time data processing. Experimenting using the UNSW-15 dataset shows that even using very small sample sizes it's possible to differentiate malicious sessions from benign ones, meaning a large advancement in AI-driven solutions to cybersecurity, by being more adept at identifying complex threats with simplified machine learning models.