Research Article
Network Intrusion Detection Based on Sparse Autoencoder and IGA-BP Network
Table 5
The comparison result of AC (%).
| Methods | G1 | G2 | G3 | G4 | G5 |
| BPNN [31] | 94.5 | 92.7 | 95.12 | 95.57 | 93.35 | PCA-PSO-BP [8] | 90.6 | 89.5 | 90.12 | 91.27 | 90.44 | PCA-ANN [6] | 94.35 | 92.7 | 95.12 | 95.57 | 93.35 | KPCA-SVM [9] | 95.4 | 94.1 | 95.8 | 96.45 | 95.32 | RNN [32] | 97.1 | 95.6 | 97.21 | 97.1 | 96.51 | DBN [33] | 97.31 | 95.01 | 97.12 | 97.57 | 96.65 | CNN-WDLSTM [34] | 97.52 | 96.04 | 97.45 | 97.71 | 97.10 | Proposed | 97.68 | 97.26 | 97.72 | 98.27 | 97.25 |
|
|