Automatic Detection of Android Malware via Hybrid Graph Neural Network

Table 4

Experimental results of different models.

Models	Binary-classify				Five-classify
	Acc	P	R	F1	Acc	P	R	F1
GCNs	98.69	98.91	99.05	98.98	98.13	97.98	98.01	98.00
Bi-GRU+GCNs	99.18	99.51	99.13	99.32	98.56	98.12	98.32	98.22
Bi-GRU+Deep_TNN +Self-attention	99.62	99.67	99.5	99.58	99.2	99.15	98.89	99.01