Table 3 The performance of our deep learning models on the NSL-KDD dataset
From: Network intrusion detection using feature fusion with deep learning
DL model | Train | Validation Performance (%) | Test |
|---|---|---|---|
Early-fusion model | Â | Â | Â |
 Accuracy | 99.74 | 99.66 | 84.50 |
 Precision | 99.76 | 99.69 | 84.98 |
 Recall | 99.72 | 99.63 | 84.31 |
Late-fusion model | Â | Â | Â |
 Accuracy | 99.50 | 99.51 | 83.73 |
 Precision | 99.57 | 99.57 | 84.01 |
 Recall | 99.41 | 99.42 | 83.52 |
Late-Ensemble model | Â | Â | Â |
 Accuracy | 99.55 | 99.43 | 86.81 |
 Precision | 99.56 | 99.46 | 86.86 |
 Recall | 99.54 | 99.41 | 86.80 |