Table 2 Performance of the proposed model after using multi-class classification architecture in terms of F1-score, FRP, Recall, Precision, and Accuracy metrics
From: Practical autoencoder based anomaly detection by using vector reconstruction error
F1-score | FPR | Recall | Precision | Accuracy | Class |
|---|---|---|---|---|---|
100 | 0 | 100 | 100 | 100 | Normal |
99.941 | 0.007 | 100 | 99.882 | 99.993 | Attacker |
99.906 | 0.007 | 100 | 99.813 | 99.992 | Victim |
99.792 | 0.118 | 100 | 99.585 | 99.907 | Unknown |
99.916 | 0 | 99.833 | 100 | 99.893 | Suspicious |