Research Article
RAEF: An Imputation Framework Based on a Gated Regulator Autoencoder for Incomplete IIoT Time-Series Data
Table 3
Imputation performance on BSS in MRE%.
| MP | MR (%) | Baseline | Ours | BM (%) | KNN | LIME-LSTM (%) | BRITS (%) | RFL (%) | RFG (%) | RFLR (%) | RFGR (%) |
| CM | 30 | 60.62 | — | 24.20 | 17.84 | 16.99 | 17.06 | 14.06 | 14.32 | 20 | 39.73 | — | 19.57 | 15.05 | 14.45 | 15.11 | 12.29 | 12.99 | 10 | 16.32 | — | 15.43 | 14.14 | 12.98 | 13.58 | 11.28 | 11.67 |
| UM | 30 | 67.44 | 47.42% | 19.11 | 15.91 | 14.12 | 14.01 | 11.97 | 12.34 | 20 | 45.23 | 31.84% | 14.28 | 13.26 | 12.28 | 12.26 | 10.94 | 11.52 | 10 | 19.01 | 16.83% | 11.85 | 11.73 | 10.88 | 11.47 | 9.94 | 10.01 |
|
|