CYBER ATTACK PREDICTION USING TRADITIONAL MACHINE LEARNING AND GENERATIVE AI MODELS

Abstract

The shift from simple machine learning to generative AI models for cyberattack prediction is examined in this research. The rapid advancement of digital technologies and the prevalence of cyber risks make it more important to anticipate and identify attacks in order to protect network infrastructures and information systems. Using out-of-date network data, several have employed classification and anomaly detection to forecast attacks. These techniques, however, don't always prevent sophisticated and evolving cyberattacks. By collecting additional data, enhancing model learning, and simulating fictitious assault scenarios, generative artificial intelligence enhances cyberattack prediction. This work enhances cybersecurity forecasts, adaptability, and proactive defense through the use of generative AI models and traditional machine learning.