A Novel Stacking Heterogeneous Ensemble Model with Hybrid Wrapper-Based Feature Selection for Reservoir Productivity Predictions

Table 5

Prediction model performances.

Model	Training set		Testing set
	IF_LR	0.8305	0.0143	0.57	0.04
IF_SVR	0.8593	0.0119	0.54	0.05
IF_NN	0.9413	0.0050	0.72	0.03
IF_RF	0.9290	0.0060	0.48	0.05
IF_Xgboost	0.9999	5.3402 ×	0.22	0.08
MF_LR	0.7901	0.0178	0.66	0.04
MF_SVR	0.9158	0.0071	0.72	0.03
MF_NN	0.9318	0.0058	0.72	0.03
MF_RF	0.9348	0.0055	0.44	0.06
MF_Xgboost	0.9999	1.1110 ×	0.14	0.09
IF_Ensemble_Average	0.9454	0.0046	0.62	0.04
IF_Ensemble_Stack	0.8254	0.0148	0.85	0.02
MF_Ensemble_Average	0.9318	0.0058	0.61	0.04
MF_Ensemble_Stack	0.9327	0.0057	0.89	0.01
IF_MF_Ensemble_Average	0.9269	0.0062	0.89	0.01
Our proposed model	0.9987	0.0001	0.91	0.01