VdaBSC: A Novel Vulnerability Detection Approach for Blockchain Smart Contract by Dynamic Analysis

Table 5

Comparing the existing methods with the proposed model on 75% training data.

Models	Measures	Reentrancy	Integer overflow	Integer underflow	TOD	URV
Oyente	F1-measure	71.82	72.45	72.58	72.67	72.79
Oyente	Accuracy	73.15	73.78	73.91	74	74.12
Oyente	Precision	74.91	75.54	75.67	75.76	75.88
Oyente	Recall	74.83	75.46	75.59	75.68	75.8
Maian	F1-measure	73.33	73.96	74.09	74.18	74.3
Maian	Accuracy	77.24	77.87	78	78.09	78.21
Maian	Precision	76.86	77.49	77.62	77.71	77.83
Maian	Recall	76.51	77.14	77.27	77.36	77.48
SmartCheck	F1-measure	77.88	78.51	78.64	78.73	78.85
SmartCheck	Accuracy	80.91	81.54	81.67	81.76	81.88
SmartCheck	Precision	79.43	80.06	80.19	80.28	80.4
SmartCheck	Recall	76.98	77.61	77.74	77.83	77.95
ContractGuard	F1-measure	83.59	84.22	84.35	84.44	84.56
ContractGuard	Accuracy	85.36	85.99	86.12	86.21	86.33
ContractGuard	Precision	82.12	82.75	82.88	82.97	83.09
ContractGuard	Recall	83.61	84.24	84.37	84.46	84.58
ContractFuzzer	F1-measure	86.91	87.54	87.67	87.76	87.88
ContractFuzzer	Accuracy	88.48	89.11	89.24	89.33	89.45
ContractFuzzer	Precision	85.73	86.36	86.49	86.58	86.7
ContractFuzzer	Recall	84.38	85.01	85.14	85.23	85.35
Proposed	F1-measure	86.58	87.21	87.34	87.43	87.55
Proposed	Accuracy	88.57	89.2	89.33	89.42	89.54
Proposed	Precision	88.73	89.36	89.49	89.58	89.7
Proposed	Recall	85.73	86.36	86.49	86.58	86.7