Hybrid attention-enhanced explainable model for encrypted traffic detection and classification
Encrypted traffic detection is critical as protocols like TLS, VPNs, and Tor dominate modern networks. We propose a Hybrid Attention–LightGBM model with an augmented multi-dataset approach and Explainable AI tools (SHAP, LIME) to enhance interpretability, scalability, and generalization. Experiments show it outperforms state-of-the-art methods in both binary and multi-class classification, advancing adaptive encrypted traffic analysis.