Research Article
A Novel Framework Design of Network Intrusion Detection Based on Machine Learning Techniques
Table 7
Evaluation results for various types of traffic under multiclass classification.
| ā | Accuracy: 0.9990 | Precision | Recall | F1-score |
| Bengin | 0.9996 | 0.9995 | 0.9995 | Bot | 0.9142 | 0.8492 | 0.8805 | DDos | 0.9999 | 0.9997 | 0.9998 | Dos goldeneye | 0.9984 | 0.9793 | 0.9887 | Dos hulk | 0.9963 | 0.9993 | 0.9978 | Dos slow http test | 0.9963 | 0.9891 | 0.9927 | Dos slow loris | 0.9960 | 0.9908 | 0.9934 | FTP-Patator | 1.0000 | 0.9983 | 0.9992 | Heartbleed | 1.0000 | 0.3333 | 0.5000 | Infiltration | 1.0000 | 0.2727 | 0.4286 | PortScan | 0.9997 | 0.9999 | 0.9998 | SSH-patator | 0.9943 | 0.9853 | 0.9898 | Web attack-brute force | 0.6808 | 0.8872 | 0.7704 | Web attack-sql injection | 1.0000 | 0.1667 | 0.2857 | Web attack-XSS | 0.4655 | 0.1378 | 0.2126 | Weighted avg | 0.9989 | 0.9990 | 0.9989 |
|
|