H. Liu et al. : Effluent Quality Prediction of Papermaking WWTPs Using SEL
TABLE 4. Specific parameters of base-learning algorithms.
FIGURE 6. Prediction results of SS eff andCOD eff using stacking ensemble learning.
using SEL. The quantitative evaluation results including RMSE, MAPE and R 2 for PLS, SVR, ANN, RF, AdaBoost and SEL are listed in Tables 5 and 6. It can be seen that
SEL achieves the highest prediction accuracy for two cases. For COD eff , compared with the base-learning algorithms of PLS, SVR, and ANN, the RMSE of SEL is reduced by
180850
VOLUME 8, 2020
Made with FlippingBook flipbook maker