Developing a Robust Model Based on the Gaussian Process Regression Approach to Predict Biodiesel Properties

Table 2

The comparison between four kernel functions for IV property from the point of view of statistical quantities.

Model	Phase	R^2	MRE (%)	MSE	RMSE	STD
GPR (Exponential)	Train	1.000	0.23024	0.15939	0.39923	0.29438
	Test	0.984	0.96971	3.15731	1.77688	1.35545
	Total	0.998	0.27729	0.79448	1.77688	0.83169
GPR (Matern)	Train	0.999	0.44408	0.44046	0.66367	0.39484
	Test	0.993	0.83689	3.02336	1.73878	1.45060
	Total	0.997	0.46640	0.98545	1.73878	0.81850
GPR (Squared Exponential)	Train	0.997	0.55285	1.18464	1.08841	0.85742
	Test	0.998	0.69364	0.93277	0.96580	0.56754
	Total	0.997	0.62869	1.22526	0.96580	0.81806
GPR (Rational Quadratic)	Train	0.996	0.60686	1.30529	1.14249	0.87268
	Test	0.996	0.74534	1.93710	1.39180	1.14923
	Total	0.996	0.64906	1.56170	1.39180	0.99212