Research Article
[Retracted] Research on Boruta-ET-Based Anomalous Traffic Detection Model
Table 6
The performance of the methods in the study.
| Method | ā | Benign | DoS | Port scan | Bot | Brute force | Web attack | Infiltration |
| NB | Precision | 0.87 | 0.81 | 0.76 | 0 | 0 | 0 | 0.01 | Recall | 0.98 | 0.58 | 0 | 0 | 0 | 0 | 0.70 | F1 | 0.92 | 0.68 | 0 | 0 | 0 | 0 | 0.02 |
| LGBM | Precision | 0.88 | 0.61 | 0.83 | 0 | 0.08 | 0.10 | 0 | Recall | 0.90 | 0.64 | 0.14 | 0 | 0.03 | 0.71 | 0 | F1 | 0.89 | 0.14 | 0.23 | 0 | 0.05 | 0.18 | 0 |
| DT | Precision | 0.99 | 0.99 | 0.99 | 0.41 | 0.95 | 0.66 | 0.17 | Recall | 1 | 0.94 | 1 | 0.63 | 0.88 | 0.93 | 1 | F1 | 0.99 | 0.97 | 1 | 00.50 | 0.91 | 0.77 | 0.29 |
| Xgboost | Precision | 0.95 | 0.99 | 1 | 0.41 | 1 | 0.99 | 0.24 | Recall | 1 | 0.93 | 0.46 | 0.25 | 0.46 | 0.86 | 1 | F1 | 0.97 | 0.96 | 0.63 | 0.31 | 0.63 | 0.92 | 0.38 |
| RF | Precision | 0.96 | 1 | 1 | 0.55 | 0.99 | 0.99 | 1 | Recall | 1 | 0.94 | 0.56 | 0.26 | 0.77 | 0.89 | 1 | F1 | 0.98 | 0.97 | 0.71 | 0.35 | 0.87 | 0.94 | 1 |
| DNN | Precision | 0.92 | 0.85 | 0.98 | 0 | 0.98 | 0 | 0 | Recall | 0.97 | 0.77 | 0.50 | 0 | 0.53 | 0 | 0 | F1 | 0.95 | 0.81 | 0.66 | 0 | 0.69 | 0 | 0 |
| The method proposed Boruta-ET | Precision | 1 | 1 | 0.99 | 0.87 | 1 | 0.98 | 1 | Recall | 1 | 1 | 1 | 0.26 | 0.98 | 0.95 | 1 | F1 | 1 | 1 | 1 | 0.40 | 0.99 | 0.97 | 1 |
|
|