A hybrid machine learning method for increasing the performance of network intrusion detection systems

Journal of Big Data

Table 6 The computational times of the proposed method

Dataset	Class	All features (s)	Feature selection (s)	Data reduction (s)
Dataset	Class	All features (s)	Feature selection (s)	std = 1	std = 2	std = 3
NSL-KDD [13]	DoS	1.72	0.231	0.475	0.3	0.32
	Probe	0.89	0.324	0.39	0.39	0.39
	R2L	0.8	0.31	0.37	0.37	0.4
	U2R	0.6	0.19	0.14	0.14 s	0.14
NSL-KDD [13]	Attack–normal	2.47	1.15	1.15	1.15	1.15
UNSW-NB15 [27]	Attack–normal	4.56	1.26	1.45	1.52	1.54