APPLYING ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING MODELS TO ENHANCE RISK ANALYSIS AND THREAT DETECTION: A COMPREHENSIVE REVIEW
Ilakiya Ulaganathan · International Journal for Research Trends and Innovation · 2025
Cyber treasures today sit on hi-tech sites; they get attacked by sophisticated methods, the incidence and complexity of attacks are on the rise: hence, traditional rule-based security systems are unable to prevent timely and accurate risk analysis and threat detection.With AI and ML as emerging technologies in cybersecurity, real-time monitoring, predictive analytics, anomaly detection, automatic response-making, and more are achievable.This review gives a thorough overview of AI and ML techniques that are applied to various domains in cybersecurity, including supervised and unsupervised learning, deep learning, reinforcement learning, and hybrid models.It studies their systems in methodologies of intrusion detection, malware classification, fraud prevention, and vulnerability assessment.The review further discusses the most used datasets, performance metrics, and limitations currently making the wide adoption slow: data quality, adversarial threat, model interpretability, and ethics.Future directions in research such as explainable AI, federated learning, and autonomous threat response systems are also discussed.This review may serve as a stepping stone for researchers and practitioners studying or utilizing AI/ML for cyber risk management in a broader sense.