Binary File’s Visualization and Entropy Features Analysis Combined with Multiple Deep Learning Networks for Malware Classification

Table 8

Proposed method compared with other malware classification methods on the malimg dataset.

Methods	Acc (%)	Recall (%)	(%)	RT (s)
GIST + KNN [3]	97.28	97.28	96.61	0.423
LBP + KNN [25]	97.87	97.88	97.86	0.052
GIST + DSIFT + RF [6]	98.53	98.53	98.49	0.026
VGG (fine-tune) [28]	98.84	98.84	98.82	59.50
Our method	99.73	99.73	99.73	7.625

RT means running time.