An Effective ABC-SVM Approach for Surface Roughness Prediction in Manufacturing Processes

Table 5

Statistical results of 30 runs obtained by ABC-SVM, DE-SVM, GA-SVM, and PSO-SVM models for TC18 titanium alloy (The values of calculated R² value are small negative values, here they are replaced by the symbol “--”).

		ABC-SVM	DE-SVM	GA-SVM	PSO-SVM
MSE	Best	0.0014	0.0014	0.0014	0.0015
	Worst	0.0014	0.0014	0.0014	0.0015
	Mean	0.0014	0.0014	0.0014	0.0015
	StdDev	0	0	0	0
R^2	Best	--	--	--	--
	Worst	--	--	--	--
	Mean	--	--	--	--
	StdDev	0	0	0	0
MAE	Best	0.0328	0.0328	0.0328	0.0298
	Worst	0.0328	0.0328	0.0328	0.0298
	Mean	0.0382	0.0328	0.0382	0.0298
	StdDev	0	0	0
T(s)		2.8574	6.3105	4.8289	3.0546