EFS-DNN: An Ensemble Feature Selection-Based Deep Learning Approach to Network Intrusion Detection System

Table 9

Five-class classification results in terms of TPR (%) and FPR (%) on KDD 99 dataset.

Method	DoS		Probe		R2L		U2R
	TPR	FPR	TPR	FPR	TPR	FPR	TPR	FPR
ID3 [27]	99.90	0.03	99.70	0.55	93.50	0.98	49.10	0.15
MARK-LEM [28]	99.96	0.02	97.30	0.03	94.94	0.05	62.87	0.01
HAST-IDS [29]	99.10	0.02	83.35	0.01	74.19	0.02	64.25	0.02
SwiftIDS [12]	99.96	0.02	98.20	0.00	89.40	0.01	57.10	0.01
EFS-DNN	99.97	0.02	98.79	0.00	98.84	0.09	66.67	0.01