An Attribute Extraction for Automated Malware Attack Classification and Detection Using Soft Computing Techniques

Table 7

Different techniques for selecting features in relation to classification techniques that use 80% of the information for trained and 20% for tests.

Attribute extraction	No. of attributes	ANN (Acc, TPR, FPR, TNR, FNR)	Decision tree (Acc, TPR,FPR, TNR, FNR)	Naive Bayes (Acc, TPR, FPR, TNR, FNR) (%)	SVM (Acc, TPR, FPR, TNR, FNR) (%)
		(%)	(%)
CFsSubset	25	95.4	97	93	98.5
		100	100	100	100
		3	6	9	3
		94	96	89	96
		0	0	0	0
PCA	55	96.4	95.8	91.5	97.7
		100	93	93	91
		4	7	9	0
		95	98	88	98
		0	7	6	8
GainRatioAttribute	201	95.7	97.3	87.9	97.5
		100	100	100	100
		6	3	13	5
		8	97	84	97
		0	0	0	0
SymmetricalUncertAttributeEval	146	94.6	94.3	90.5	97.9
		100	100	100	100
		5	6	12	3
		94	93	85	96
		0	0	0	0