fig5

Figure 5. Changes in OC during iteration across different datasets: (A-C) Ni-based dataset with (A) SVR, (B) XGBoost, and (C) NN; (D-F) Fe-based dataset with (D) SVR, (E) XGBoost, and (F) NN; (G-I) Ti-based dataset with (G) SVR, (H) XGBoost, and (I) NN. In panels (D) and (G), the two curves overlap completely, indicating identical OC evolution behaviors. OC: Opportunity cost; SVR: support vector regression; XGBoost: extreme gradient boosting; NN: neural network.