Table 4 Performance metrics of ET-SSL on benchmark datasets.
From: Anomaly detection in encrypted network traffic using self-supervised learning
Dataset | Accuracy (%) | Precision (%) | Recall (%) | F1-score (%) | FPR (%) | Throughput (flows/s) |
|---|---|---|---|---|---|---|
CIC-Darknet2020 | 96.8 | 95.5 | 94.3 | 94.9 | 1.2 | 1500 |
ISCX VPN-nonVPN | 94.3 | 92.1 | 93.7 | 92.9 | 2.4 | 1350 |
UNSW-NB15 | 95.1 | 93.9 | 94.8 | 94.3 | 1.7 | 1450 |
