Gradient Boosted Trees Predictive Models for Surface Roughness in High-Speed Milling in the Steel and Aluminum Metalworking Industry

Table 17

Comparison between models in each dataset.

Dataset	GBT	SVM	Random Forests	p-value
Slots M1	Accuracy: 83.20% +- 6.27%; (40, 4, 5)	Accuracy: 80.20% +- 5.02%; (RBF, 10, 0.10)	Accuracy: 87.60% +- 6.64%; (80, 5)	0.2746
Slots M2	Accuracy: 72.69% +- 7.19%; (60, 6, 4)	Accuracy: 75.77% +- 4.88%; (RBF, 10, 1.0)	Accuracy: 77.31% +- 9.80%; (30, 8)	0.4308
Geometries M1	Accuracy: 85.29% +- 2.56%; (100, 8, 10)	Accuracy: 82.76% +- 2.47%; (RBF, 1000, 1.0)	Accuracy: 86.44% +- 3.20%; (30, 8)	0.0263
Geometries M2	Accuracy: 83.68% +- 4.98%; (40, 4, 8)	Accuracy: 83.27% +- 4.64%; (RBF, 1000, 1.0)	Accuracy: 86.12% +- 4.26%; (20, 8)	0.3817