An Investigation into AI-driven real-time Network Traffic Classification and Anomaly Detection in Complex Network Environments
Xinxia Li · 2025
This study investigates the use of artificial intelligence (AI) in network traffic classification and anomaly detection, and the usefulness of such methods over traditional techniques. The research employs a metaanalysis approach to collect and analyse findings from peer reviewed studies to draw conclusions about the accuracy, scalability, and adaptability of AI in dynamic network environments. The results demonstrate that AI surpasses the conventional techniques for both anomaly detection and classification but still retain computational inefficiencies, adversarial vulnerabilities and scalability limitations. The study emphasizes data privacy and ethical considerations when deploying AI, and provides insights to strengthen further the role of AI in network security.