Research Article
Towards Effective Network Intrusion Detection: A Hybrid Model Integrating Gini Index and GBDT with PSO
Table 7
Performance comparison of various methods in detection of abnormal connections.
| Method | Accuracy | Precision | DR | F1-score | FAR |
| SVM | 73.41% | 96.89% | 66.28% | 78.72% | 2.81% | RF | 76.57% | 96.86% | 60.83% | 74.73% | 2.61% | C4.5 | 80.58% | 95.08% | 69.49% | 80.30% | 4.75% | FC-ANN | 75.80% | 96.95% | 59.36% | 73.64% | 2.47% | CFA-DT | 72.24% | 93.42% | 73.58% | 82.32% | 3.45% | IGCR-ANN | 77.83% | 96.85% | 63.10% | 76.41% | 2.71% | GINI-GBDT-PSO | 86.10% | 96.44% | 78.48% | 86.54% | 3.83% |
|
|