Network traffic anomaly detection based on deep learning: a review

Network traffic anomaly detection has become an important research topic with the increasing prevalence of network attacks. Deep learning, with its ability to analyse large-scale datasets, has emerged as a powerful tool for network traffic anomaly detection. This paper presents a comprehensive overview of state-of-the-art deep learning-based network traffic anomaly detection models including VAE, BiLSTM, and vision transformer, in terms of dimensional deduction, time dependence and data imbalance. The performance of these models has been evaluated and compared on KDDCUP99 and CICIDS2017 datasets. Finally, we outline challenges and future research aimed at enhancing the performance and practicality of network traffic anomaly detection based on deep learning.