Table 7 Comparative performance metrics over N_BaIoT and UNSW-NB15 datasets.
From: Securing IoT networks: a machine learning approach for detecting unusual traffic patterns
Model type | Dataset | Accuracy (± SD) | Precision (± SD) | Recall (± SD) | F1 Score (± SD) |
|---|---|---|---|---|---|
Decision Tree | N_BaIoT | 91.0% ± 1.2 | 90.5% ± 1.3 | 89.7% ± 1.4 | 90.1% ± 1.2 |
UNSW-NB15 | 94.0% ± 1.1 | 93.5% ± 1.2 | 92.8% ± 1.3 | 93.1% ± 1.1 | |
SVM | N_BaIoT | 93.5% ± 1.0 | 92.8% ± 1.1 | 93.0% ± 1.0 | 92.9% ± 1.1 |
UNSW-NB15 | 96.0% ± 0.9 | 95.0% ± 1.0 | 94.5% ± 1.0 | 94.7% ± 0.9 | |
Random Forest | N_BaIoT | 95.0% ± 0.8 | 94.2% ± 0.9 | 94.5% ± 0.8 | 94.3% ± 0.9 |
UNSW-NB15 | 97.0% ± 0.7 | 96.5% ± 0.8 | 96.2% ± 0.7 | 96.4% ± 0.7 | |
Neural Network | N_BaIoT | 97.2% ± 0.6 | 96.8% ± 0.6 | 97.5% ± 0.5 | 97.1% ± 0.6 |
UNSW-NB15 | 98.7% ± 0.4 | 98.3% ± 0.5 | 99.0% ± 0.3 | 98.6% ± 0.4 |
