Network Traffic Classification Based on SD Sampling and Hierarchical Ensemble Learning

Table 12

Classification performance of the sampled dataset with the SD sampling algorithm.

	Precision	Recall	F1 score	FNR	FPR
SD sampling layer1
Benign	0.9996	0.9972	0.9984	0.0004	0.0028
Abnormal	0.9972	0.9996	0.9984	0.0028	0.0004
Accuracy					0.9984
Macro avg	0.9984	0.9984	0.9984	0.0016	0.0016
Weighted avg	0.9984	0.9984	0.9984	0.0016	0.0016
SD sampling layer2
Benign	0.9996	0.9972	0.9984	0.0004	0.0028
DoS hulk	0.9811	0.9988	0.9899	0.0189	0.0012
DDoS	1.0000	1.0000	1.0000	0.0000	0.0000
PortScan	1.0000	0.9996	0.9998	0.0000	0.0004
DoS goldeneye	0.9980	0.9988	0.9984	0.0020	0.0012
FTP-patator	1.0000	1.0000	1.0000	0.0000	0.0000
DoS slowloris	0.9955	0.9911	0.9933	0.0045	0.0089
DoS slowhttptest	0.9962	0.9908	0.9935	0.0038	0.0092
SSH-patator	1.0000	1.0000	1.0000	0.0000	0.0000
Bot	1.0000	0.9959	0.9979	0.0000	0.0041
Infiltration	1.0000	1.0000	1.0000	0.0000	0.0000
Heartbleed	1.0000	1.0000	1.0000	0.0000	0.0000
Web attack	0.9853	0.9981	0.9917	0.0147	0.0019
Accuracy					0.9975
Macro avg	0.9966	0.9977	0.9971	0.0034	0.0023
Weighted avg	0.9976	0.9975	0.9975	0.0024	0.0025