Table 1 Comparison of the predictive accuracies and lack-of-fit statistics for four models—PQM, ANN, LLR, and GPR

From: Unveiling the effect of composition on nuclear waste immobilization glasses’ durability by nonparametric machine learning

  

PQM

ANN

LLR

GPR

Training set

R2

0.62 ± 0.01

0.78 ± 0.01

0.77 ± 0.18

0.76 ± 0.01

RMSE [ln(*)]

1.05 ± j0.01

0.80 ± 0.02

0.66 ± 0.16

0.83 ± 0.02

Test set

R2

0.58 ± 0.04

0.63 ± 0.08

0.61 ± 0.15

0.65 ± 0.07

RMSE [ln(*)]

1.09 ± 0.05

1.03 ± 0.12

0.93 ± 0.23

1.00 ± 0.12

Lack of fit, F

1.50 ± 0.00

0.87 ± 0.03

1.28 ± 0.01

0.63 ± 0.04

  1. The mean and standard deviation values are calculated based on the model performance over ten train-test splits. Here, * represents the unit of the scaled VHT alteration rate, i.e., g/(m2·d). Lack of fit statistics are calculated based on all samples in the dataset.
Back to article page