Research Article
A Deep Learning Anomaly Detection Framework for Satellite Telemetry with Fake Anomalies
Table 3
Comparison of LSTMs with LSTMs-DDMN in precision, recall, and F1-score.
| | Precision | Recall | F1-score |
| 2 | 0.994 | 0.994 | 0.968 | 0.997 | 0.981 | 0.995 | 2.5 | 0.998 | 0.998 | 0.963 | 0.986 | 0.975 | 0.992 | 3 | 0.999 | 0.999 | 0.914 | 0.957 | 0.955 | 0.978 | 3.5 | 0.999 | 0.999 | 0.761 | 0.892 | 0.864 | 0.943 | 4 | 0.999 | 0.999 | 0.553 | 0.825 | 0.713 | 0.904 | | LSTMs | LSTMs-DDMN | LSTMs | LSTMs-DDMN | LSTMs | LSTMs-DDMN |
|
|