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

Table 1

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

Model	Phase	R^2	MRE (%)	MSE	RMSE	STD
GPR (Exponential)	Train	0.998	2.61001	0.00622	0.07884	0.06408
	Test	0.975	4.50131	0.12657	0.35577	0.29305
	Total	0.991	3.25054	0.03573	0.35577	0.16650
GPR (Matern)	Train	0.994	1.19836	0.02510	0.15841	0.12080
	Test	0.992	4.29964	0.02464	0.15697	0.08520
	Total	0.993	2.04742	0.02624	0.15697	0.11351
GPR (Squared Exponential)	Train	0.993	2.40750	0.01830	0.13529	0.10158
	Test	0.981	5.07691	0.13892	0.37272	0.28103
	Total	0.991	2.38574	0.03773	0.37272	0.16620
GPR (Rational Quadratic)	Train	0.994	2.77862	0.02566	0.16019	0.12024
	Test	0.969	6.90077	0.08160	0.28565	0.14865
	Total	0.991	3.50671	0.03681	0.28565	0.14139