Evaluating Drug Risk Using GAN and SMOTE Based on CFDA’s Spontaneous Reporting Data

Table 4

Evaluation metrics based on the test set.

Classifier	Label	Precision	Recall	F1	Accuracy (%)
Model 1 (RF)	0	0.85	0.99	0.92
	1	0.50	0.03	0.06	84.44
	2	0.00	0.00	0.00
	Macro-avg	0.45	0.34	0.32
	Weighted-avg	0.78	0.84	0.78
Model 2 (SMOTE + RF)	0	0.91	0.85	0.88
	1	0.91	0.94	0.92	91.99
	2	0.94	0.97	0.96
	Macro-avg	0.92	0.92	0.92
	Weighted-avg	0.92	0.92	0.92
Model 3 (FS_GAN + SMOTE + RF)	0	0.93	0.96	0.94
	1	0.98	0.94	0.96	96.25
	2	0.98	0.98	0.98
	Macro-avg	0.96	0.96	0.96
	Weighted-avg	0.96	0.96	0.96