A novel time efficient learning-based approach for smart intrusion detection system

Journal of Big Data

Table 10 Comparison of the Proposed model with other machine learning algorithms

	Feature selection	Prediction latency	Accuracy (%)	Recall (%)	Precision (%)	Specificity (%)	F-measure (%)
Histogram Gradient Boosting	PCA	0.387566	97.64	96.19	98.93	99.02	97.54
Histogram Gradient Boosting	Hybrid	0.300503	97.80	96.04	99.36	99.43	97.67
ExtraTrees	PCA	0.353288	92.81	93.77	91.72	91.89	92.74
ExtraTrees	Hybrid	0.345334	91.98	93.24	90.26	90.85	91.72
RandomForest	PCA	0.352546	97.65	95.99	99.16	99.2	97.55
RandomForest	Hybrid	0.33835	97.34	95.67	98.74	98.88	97.18
XGBoost	PCA	0.329168	97.50	95.61	99.24	99.30	97.39
XGBoost	Hybrid	0.182613	96.97	94.79	98.82	98.96	96.77
KNN	PCA	310.8692	97.47	96.30	98.49	98.59	97.38
KNN	Hybrid	186.4265	97.68	96.16	98.96	99.0	97.54
Light GBM + Hybrid feature selection	PCA	0.143057	97.82	96.12	99.40	99.44	97.73
Light GBM + Hybrid feature selection	hybrid	0.138008	97.72	96.06	99.33	99.42	97.57