A Review of AI-Based Network Anomaly Traffic Detection

In the face of increasingly complex and covert cyberattacks, cybersecurity has become paramount, yet traditional methods face limitations. Consequently, AI-driven anomaly traffic detection has emerged as a hot research topic in the field. This review systematically organizes research progress in AI-based network anomaly traffic detection. It first defines anomaly traffic types and analyzes limitations of traditional detection methods, then focuses on key technologies of machine learning and deep learning models in feature learning and anomaly identification. The review compares the advantages and disadvantages of different models, finally analyzing current challenges and outlining future research directions.