Cybersecurity

Table 3 Results for 2 CWE types

From: Graph neural network based approach to automatically assigning common weakness enumeration identifiers for vulnerabilities

Model	P	R	F1	A
SVM	0.78	0.47	0.59	0.657
Naïve Bayes	0.56	0.69	0.62	0.658
CNN	0.79	0.68	0.73	0.853
LSTM	0.94	0.78	0.85	0.895
GCN	0.74	0.89	0.81	0.926
Our method	0.83	0.94	0.88	0.953

Back to article page