AI-Driven Anomaly Detection in Network Security: A Comparative Study of Machine Learning Algorithms

Abstract

As cyber threats continue to grow in sophistication and frequency, traditional network security measures often struggle to detect novel and evolving attacks. Anomaly detection, driven by Artificial Intelligence (AI) and Machine Learning (ML), has emerged as a powerful technique for identifying abnormal patterns in network traffic that may indicate security breaches. a comparative study of various machine learning algorithms used in anomaly detection for network security. Specifically, we examine supervised, unsupervised, and deep learning models, including decision trees, k-means clustering, support vector machines (SVM), and neural networks, evaluating their effectiveness in detecting anomalies and mitigating cyber threats. Through simulations and real-world data analysis, the study highlights the strengths and limitations of each algorithm in terms of detection accuracy, false positives, computational complexity, and adaptability to different network environments. We also discuss the challenges of deploying AI-driven anomaly detection systems in practice, including data quality, model scalability, and the risk of adversarial attacks. This comparative study aims to provide insights into the most effective ML algorithms for enhancing network security, offering recommendations for future research and practical implementations in AI-driven cybersecurity solutions.