LNNLS-KH: A Feature Selection Method for Network Intrusion Detection

Table 16

The classification FPR and DR of different feature selection algorithms (CICIDS2017 dataset).

Data categories	False positive rate (FPR) (%)					Detection rate (DR) (%)
	CMPSO	ACO	KH	IKH	LNNLS-KH	CMPSO	ACO	KH	IKH	LNNLS-KH
Normal	9.25	8.72	6.41	4.93	3.67	88.05	88.51	89.25	92.46	93.89
DoS	5.41	4.48	4.06	2.83	1.94	72.57	82.89	87.86	92.56	92.64
DDoS	6.85	4.92	4.54	6.33	3.18	79.03	83.47	90.22	87.52	92.98
PortScan	4.65	3.02	2.84	1.86	1.16	88.25	93.80	94.33	95.14	95.42
WebAttack	5.33	3.16	2.52	2.11	1.60	87.40	91.35	92.19	92.94	94.77