Table 5 Quantitative performance indicators of models using aggregated data.
Final test results (Aggregated composite data) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
Model | Training | Testing | ||||||||||
MAE | RMSE | R2 | 95th percentile error | 99th percentile error | Max_error | MAE | RMSE | R2 | 95th percentile error | 99th percentile error | Max_error | |
Decision tree | 0.7276 | 0.9032 | 0.9975 | 1.7733 | 2.1496 | 2.3283 | 1.1732 | 1.5914 | 0.9913 | 3.4575 | 3.7499 | 3.8150 |
LightGBM | 0.7247 | 0.9041 | 0.9975 | 1.8207 | 2.1463 | 2.2699 | 1.1683 | 1.5941 | 0.9913 | 3.4623 | 3.7551 | 3.8202 |
Random forest | 0.7349 | 0.9115 | 0.9975 | 1.7680 | 2.0993 | 2.3577 | 1.2493 | 1.6363 | 0.9908 | 3.5792 | 3.8788 | 3.9439 |
Polynomial MLR (deg = 2) | 1.0225 | 1.2883 | 0.9949 | 2.3979 | 2.9620 | 3.6508 | 1.8854 | 2.2251 | 0.9830 | 3.4555 | 3.4822 | 3.4879 |
Ridge | 3.3034 | 3.9182 | 0.9531 | 7.1056 | 7.8891 | 8.6654 | 2.5568 | 3.0824 | 0.9675 | 5.0352 | 5.7710 | 5.9561 |
MLR | 3.3152 | 3.9067 | 0.9534 | 6.8441 | 7.5090 | 8.5545 | 2.6536 | 3.1926 | 0.9651 | 5.1591 | 6.0281 | 6.2501 |
