A Smart Contract Vulnerability Detection Model Based on Syntactic and Semantic Fusion Learning

Table 2

Dataset experimental results.

	Methods	Accuracy	Precision	Recall	F1
ImplicitVisibility	Conkas	—	—	—	—
	Opcode	54.16	61.86	21.73	32.16
	Our model	89.00	87.58	92.04	89.75
IntegerOverflow	Conkas	56.63	24.64	87.51	38.45
	Opcode	83.28	81.71	85.75	83.68
	Our model	95.58	94.08	97.79	95.90
IntegerUnderflow	Conkas	87.01	63.09	72.76	67.58
	Opcode	72.70	71.61	75.20	75.48
	Our model	96.42	92.86	88.51	90.63
TimeDependency	Conkas	92.97	57.64	67.49	62.18
	Opcode	94.04	97.01	90.89	93.85
	Our model	98.25	97.52	94.85	94.81
Reentrancy	Conkas	83.02	49.80	70.22	58.27
	Opcode	87.51	81.48	97.08	88.60
	Our model	98.64	90.91	78.38	75.36