Table 8.
Performance of the IDS models (EnsHMV and BoostedEnsML) in detecting and classifying each network traffic class in the two datasets.
| EnsHMV | BoostedEnML | ||||||
|---|---|---|---|---|---|---|---|
| Dataset | Class | Precision | Recall | F-score | Precision | Recall | F-Score |
| CICIDS2017 | Benign | 0.9795 | 0.9945 | 0.9636 | 0.9989 | 0.9995 | 1.0000 |
| Bot | 0.9977 | 0.9992 | 0.9984 | 0.9997 | 0.9999 | 0.9999 | |
| Brute Force | 0.9998 | 0.9999 | 0.9999 | 1.0000 | 1.0000 | 0.9999 | |
| DDoS | 0.9880 | 0.9989 | 0.9889 | 0.9890 | 0.9965 | 0.9980 | |
| DoS | 0.9968 | 0.9890 | 0.9969 | 1.0000 | 1.0000 | 1.0000 | |
| Hearbleed | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | |
| Infiltration | 0.9989 | 0.9967 | 0.9969 | 1.0000 | 1.0000 | 1.0000 | |
| PortScan | 0.9993 | 0.9992 | 0.9995 | 0.9999 | 0.9999 | 0.9999 | |
| Web Attack | 0.9966 | 0.9966 | 0.9988 | 1.0000 | 1.0000 | 1.0000 | |
| Precision | Recall | F-score | Precision | Recall | F-Score | ||
| CICIDS2018 | Benign | 0.9966 | 0.9990 | 0.9978 | 0.9999 | 0.9999 | 0.9998 |
| Bot | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | |
| Brute Force | 0.9999 | 1.0000 | 1.0000 | 1.0000 | 0.9999 | 0.9999 | |
| DDoS | 0.9999 | 1.0000 | 1.0000 | 1.0000 | 0.9999 | 1.0000 | |
| DoS | 0.9999 | 0.9999 | 0.9999 | 0.9999 | 1.0000 | 0.9999 | |
| Infiltration | 0.9999 | 1.0000 | 0.9999 | 1.0000 | 1.0000 | 1.0000 | |
| Web Attack | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | 1.0000 | |