Table 2

Performance comparison of baseline models

	Model_H	LR_H	RF_H	SVR_H
MSE	0.1395 (0.1311, 0.1479)	0.2051 (0.1926, 0.2176)	0.1614 (0.1498, 0.1730)	0.1544 (0.1436, 0.1652)
R²	0.8561 (0.8484, 0.8638)	0.7882 (0.7759, 0.8005)	0.8336 (0.8233, 0.8439)	0.8410 (0.8317, 0.8503)
Pearson	0.9258 (0.9218, 0.9298)	0.8922 (0.8860, 0.8984)	0.9142 (0.9087, 0.9197)	0.9182 (0.9134, 0.9230)

MSE, mean squared error; R², coefficient of determination; Model_H, herbal perturbation prediction model; LR_H, logistic regression model; RF_H, random forest regression model; SVR_H, support vector regression model. Bold values denote the best-performing results for each metric.