Table 3 Performance metrics for each constructed predictive framework, focused on mass density, were derived across the development, evaluation, and verification stages.

Model	R2			RMSE			AARE%
Model	Training	Test	Total	Training	Test	Total	Training	Test	Total
KNN	0.998095014	0.996494995	0.997755089	0.315177263	0.442808054	0.34479066	0.024375097	0.035191974	0.026560324
Ensemble learning	0.992627854	0.990271044	0.992081366	0.61914005	0.749935439	0.647695846	0.048820562	0.062278063	0.051539249
CNN	0.99506545	0.993781881	0.994611999	0.506785047	0.639032231	0.53613751	0.039454475	0.052329489	0.042055488
AdaBoost	0.991807639	0.987025464	0.990736994	0.652656774	0.864340544	0.700595705	0.051857955	0.07410299	0.056351902
MLP-ANN	0.998202949	0.996647242	0.997800542	0.307410834	0.463221685	0.344613659	0.02314919	0.034362955	0.025414597

Quick links

Search