LEVERAGING AI AND MACHINE LEARNING FOR PREDICTING, DETECTING, AND MITIGATING CYBERSECURITY THREATS: A COMPARATIVE STUDY OF ADVANCED MODELS
Abstract
This study investigates the use of artificial intelligence (AI) and machine learning (ML) models to predict, detect, and mitigate cybersecurity threats, including zero-day attacks, ransomware, and insider threats. Using a comprehensive dataset of network logs and attack signatures, we evaluated models such as Logistic Regression, Random Forest, XGBoost, CNN, and LSTMOur results demonstrate that deep learning models, particularly CNN (97.3% AUC-ROC) and LSTM (96.8% AUC-ROC), significantly outperform traditional methods, excelling in real-time threat detection and minimizing false positives. This study highlights the practical applicability of AI and ML in enhancing cybersecurity frameworks, paving the way for more efficient and scalable solutions against evolving threats.