Research Article
RETRACTED: An Autonomous Cyber-Physical Anomaly Detection System Based on Unsupervised Disentangled Representation Learning
| Algorithm | MSE | Accuracy (%) | Precision | Recall | r-error |
| TDVA | 0.0123 | 96.53 | 0.973 | 0.976 | 0.169 | Variational autoencoder | 0.0199 | 93.86 | 0.941 | 0.941 | 0.242 | Denoising autoencoder | 0.0259 | 90.64 | 0.902 | 0.905 | 0.902 | Sparse autoencoder | 0.0294 | 88.71 | 0.881 | 0.882 | 0.301 | Convolutional autoencoder | 0.0287 | 89.17 | 0.906 | 0.893 | 0.296 | Contractive autoencoder | 0.0182 | 93.98 | 0.945 | 0.944 | 0.238 |
|
|