TABLE 2.

Performance metrics of the assessed Machine Learning algorithms.

Model	MSE train	MSE test	MSE validation	RMSE train	RMSE test	RMSE validation	R2 train	R2 test	R2 validation
Lasso-	0.805	0.953	0.808	0.648	0.908	0.653	0.463	0.355	0.453
Lasso+	0.624	0.657	0.597	0.389	0.432	0.356	0.702	0.679	0.723
GR-	0.243	0.388	0.329	0.059	0.150	0.108	0.951	0.894	0.910
GR+	0.231	0.426	0.325	0.054	0.181	0.105	0.959	0.865	0.918
XGB-	0.186	0.308	0.254	0.035	0.095	0.065	0.971	0.933	0.946
XGB+	0.179	0.367	0.288	0.032	0.135	0.083	0.992	0.917	0.914
RF-	0.134	0.310	0.291	0.018	0.096	0.084	0.985	0.933	0.927
RF+	0.137	0.394	0.324	0.019	0.155	0.105	0.986	0.884	0.918
SVR-	0.273	0.378	0.251	0.074	0.143	0.063	0.937	0.905	0.953
SVR+	0.233	0.400	0.290	0.054	0.160	0.084	0.958	0.881	0.934
ANN-	0.271	0.381	0.260	0.074	0.145	0.067	0.939	0.897	0.944
ANN+	0.210	0.534	0.309	0.044	0.285	0.095	0.966	0.788	0.926

Abbreviations: +, including IDAC, as variable; -, excluding IDAC, as variable; GR, gaussian regression; XGB, XG, boost; RF, random forest; SVR, support vector regressor; ANN, artificial neural network.