Table 3. Model performance of LR and ML... | Emerald Publishing

Table 3.

Model performance of LR and ML models for datasets D1 and D2

Models	Dataset*	Metrics
Models	Dataset*	MAE	RMSE	R²
XGBoost	D1	4.713	6.264	0.099
XGBoost	D2	4.543	6.021	0.168
AdaBoost	D1	4.772	6.337	0.080
AdaBoost	D2	4.619	6.134	0.136
CatBoost	D1	4.725	6.298	0.088
CatBoost	D2	4.700	6.167	0.126
Extra tree (ET)	D1	4.733	6.350	0.081
Extra tree (ET)	D2	4.539	6.059	0.157
Gradient boosting (GB)	D1	4.724	6.296	0.090
Gradient boosting (GB)	D2	4.656	6.092	0.148
Random Forest (RF)	D1	4.751	6.335	0.082
Random Forest (RF)	D2	4.588	6.096	0.147
Linear regression (LR)	D1	4.843	6.337	0.078
Linear regression (LR)	D2	4.882	6.327	0.081

Models	Dataset*	Metrics
Models	Dataset*	MAE	RMSE	R²
XGBoost	D1	4.713	6.264	0.099
XGBoost	D2	4.543	6.021	0.168
AdaBoost	D1	4.772	6.337	0.080
AdaBoost	D2	4.619	6.134	0.136
CatBoost	D1	4.725	6.298	0.088
CatBoost	D2	4.700	6.167	0.126
Extra tree (ET)	D1	4.733	6.350	0.081
Extra tree (ET)	D2	4.539	6.059	0.157
Gradient boosting (GB)	D1	4.724	6.296	0.090
Gradient boosting (GB)	D2	4.656	6.092	0.148
Random Forest (RF)	D1	4.751	6.335	0.082
Random Forest (RF)	D2	4.588	6.096	0.147
Linear regression (LR)	D1	4.843	6.337	0.078
Linear regression (LR)	D2	4.882	6.327	0.081

Note(s):

* D1 contains project characteristic factors only, D2 contains both project characteristic and economic factors

Source(s): Authors’ own work