Fig. 1. Results of the PCC analysis

Fig. 2. Illustrations of performance of the six ML models in predicting K_d

Fig. 3. Distributions of normalized RMSE and MAE values by Monte Carlo cross-validation (a) RMSE, (b) MAE (color online: the central white dot denotes the median, the thick black bar indicates the interquartile range, and the black line represents the 95% confidence interval)

Fig. 4. Generalization performance of the RF model on test set

Fig. 5. Probability density distribution of K_d (a) Experimental K_d , (b) MC predicted K_d , (c) LHS predicted K_d

Fig. 6. Illustration of relative importance of input factors in RF model (a) and SHAP analysis (b) (color online)

Fig. 7. Influence of individual factors on K_d (a) pH, (b) Initial concentration (C_init), (c) Ionic strength, (d) Temperature