Fig. 4: Evaluation of predictive model performance for pollutant removal.
From: Ecological levers for microbially driven water treatment enhance pollutant removal prediction
a Comparative model performance of 15 machine learning algorithms across 4 microbial feature groups (AB, AC, BC, ABC), evaluated using mean absolute error (MAE), mean absolute percentage error (MAPE), mean squared error (MSE), root mean squared error (RMSE), and R-squared (R2). Color intensity represents the magnitude of each evaluation indicators. b Performance of machine learning algorithms using ABC input for predicting pollutant removal rates. Blue points represent training samples, and orange points represent test samples. RF random forest, ET extra trees, LightGBM light gradient boosting machine, AdaBoost adaptive boosting, XGBoost extreme gradient boosting, CatBoost categorical boosting, GBR gradient boosting regression, SVR support vector regression, Lasso least absolute shrinkage and selection operator, Bay bayesian ridge, ARD automatic relevance determination, KNN K-nearest neighbor, ELM extreme learning machine, KAN kolmogorov-arnold network, DT decision trees.
