Research Article
Intrusion Detection for Cyber-Physical Security System Using Long Short-Term Memory Model
Table 2
Comparison with the state of arts for multiclass classification.
| References | Method | Accuracy (%) | Precision (%) | Recall (%) | F1-score (%) |
| Hou et al. [8] | HLSTM-IDS | 83.85 | — | — | — | Yin et al. [10] | RNN | 81.29 | — | — | — | Ait Tchakoucht and Ezziyyani [11] | RNN-MLESM | 81.00 | — | — | — | Shone et al. [12] | S-NDAE | 85.42 | 100.00 | 85.42 | 87.37 | Le et al. [13] | LSTM | 92.00 | — | — | — | Imrana et al. [14] | BiDLSTM | 91.36 | 92.81 | 91.36 | 91.67 | Kunang et al. [15] | DAE + DNN | 83.83 | 86.02 | 83.83 | 82.04 | Imrana et al. [17] | Chi-square BidLSTM | 95.62 | — | — | 95.65 | This work | Multilayer LSTM | 96.00 | 83.00 | 100.00 | 93.00 |
|
|