From: Network traffic classification model based on attention mechanism and spatiotemporal features
Method | VPN | Non-VPN | ||||
---|---|---|---|---|---|---|
Precision (\(\mathbf{10}^{\boldsymbol-\mathbf2}\)) | Recall (\(\mathbf{10}^{\boldsymbol-\mathbf2}\)) | F1_score (\(\mathbf{10}^{\boldsymbol-\mathbf2}\)) | Precision (\(\mathbf{10}^{\boldsymbol-\mathbf2}\)) | Recall (\(\mathbf{10}^{\boldsymbol-\mathbf2}\)) | F1_score (\(\mathbf{10}^{\boldsymbol-\mathbf2}\)) | |
C4.5 [10] | 78.2 | 81.3 | - | 84.3 | 79.3 | - |
ID CNN [1] | 92 | 95.2 | - | 85.8 | 85.9 | - |
SAE + 1D CNN [11] | 97.8 | 96.3 | 97 | 86.7 | 88.8 | 87.3 |
Our | 99.4 | 99.4 | 99.4 | 97.4 | 97.5 | 96.8 |