Detecting Portable Executable Malware by Binary Code Using an Artificial Evolutionary Fuzzy LSTM Immune System

Table 2

Performance metrics.

Correctly classified instances										19344	98.59%
Incorrectly classified instances										276	01.41%
Weighted average
Method	Accuracy	TP rate	Precision	Recall	F1-score	MCC	ROC area	PRC area	MAE	RMSE	K stats
Proposed	98.59%	0.986	0.986	0.986	0.986	0.971	0.987	0.981	0.0141	0.1167	0.9714
SVM	92.91%	0.929	0.930	0.930	0.930	0.935	0.960	0.965	0.0205	0.1233	0.9398
NaBayes	88.38%	0.884	0.884	0.885	0.885	0.884	0.890	0.890	0.0318	0.2034	0.8904
k-NN	90.63%	0.906	0.900	0.900	0.910	0.900	0.900	0.950	0.0297	0.1293	0.9008
RF	97.03%	0.970	0.970	0.970	0.970	0.965	0.970	0.972	0.0170	0.1195	0.9682

SVM = support vector machines; NaBayes = naïve Bayes k-NN = k nearest neighbor; RF = random forest; TP rate = true positive rate; FP rate = false positive rate; MCC = Matthews correlation coefficient; ROC area = receiver operating characteristic area; PRC area = Precision-Recall curve area; MAE = mean absolute error; RMSE = root mean square error.