A Malware Detection Scheme Based on Mining Format Information

Table 6

Experimental results for all classification algorithms (the datasets are divided into the training set and the test set in chronological order).

Feature selection method	The number of features	Algorithm	TPR (%)	FPR (%)	Accuracy (%)	AUC
Filter	20	J48	90.7	2.1	94.6	0.928
		Random forest	94.7	1.5	96.7	0.993
		Adaboost (J48)	93.1	1.4	96.0	0.993
		Bagging (J48)	91.3	1.5	95.1	0.969
Wrapper	20	Random forest	96.5	1.3	97.6	0.994