Application of a New Feature Generation Algorithm in Intrusion Detection System

Table 7

Multiclassification results on the NSL-KDD dataset.

	Precision	Recall	F1-score	Support
nmap	1.0000	1.0000	1.0000	73
smurf	1.0000	1.0000	1.0000	665
perl	0.2000	1.0000	0.3333	2
Normal	1.0000	1.0000	1.0000	9711
satan	0.9986	1.0000	0.9993	735
guess_passwd	0.4861	0.3826	0.4282	1231
pod	0.9762	1.0000	0.9880	41
warezmaster	0.0000	0.0000	0.0000	944
teardrop	1.0000	1.0000	1.0000	12
ftp_write	0.0000	0.0000	0.0000	3
rootkit	0.6429	0.6923	0.6667	13
back	0.2297	1.0000	0.3736	359
ipsweep	0.9929	0.9858	0.9893	141
buffer_overflow	1.0000	0.7000	0.8235	20
portsweep	0.9937	1.0000	0.9968	157
phf	1.0000	0.5000	0.6667	2
land	1.0000	1.0000	1.0000	7
imap	0.0000	0.0000	0.0000	1
neptune	1.0000	1.0000	1.0000	4657
multihop	1.0000	1.0000	1.0000	18
loadmodule	1.0000	1.0000	1.0000	2
Accuracy			0.9084	18794
Macro avg.	0.7390	0.7743	0.7269	18794
Weighted avg.	0.9006	0.9084	0.8995	18794