A Novel Malware Detection and Family Classification Scheme for IoT Based on DEAM and DenseNet

Table 9

Precision, recall, and F1 score of our model on BIG 2015 dataset.

	Ramnit	Lollipop	Kelihos_ver3	Vundo	Simda	Tracur	Kelihos_ver1	Obfuscator.ACY	Gatak	Precision	Recall	F1 score
Ramnit	295	2	0	0	0	4	1	6	1	0.973	0.954	0.964
Lollipop	0	491	0	0	0	0	0	2	3	0.985	0.989	0.987
Kelihos_ver3	0	0	588	0	0	0	0	0	0	0.998	1.000	0.999
Vundo	0	1	0	91	0	1	1	1	0	0.957	0.957	0.957
Simda	0	0	0	1	7	0	0	0	0	0.875	0.875	0.875
Tracur	2	2	0	1	0	140	1	0	4	0.945	0.933	0.939
Kelihos_ver1	0	0	0	0	0	0	76	0	3	0.950	0.962	0.955
Obfuscator.ACY	6	1	1	2	1	3	1	227	3	0.957	0.926	0.941
Gatak	0	1	0	0	0	0	0	1	200	0.934	0.990	0.961
Macro										0.953	0.954	0.954