Gaussian Process Regression Tuned by Bayesian Optimization for Seawater Intrusion Prediction

Table 4

ANOVA outcomes.

Source	Sum sq.	d.f.	Mean sq.	F	Value
‘Regressor’	306492.5	4	76623.113	122442.294	0.00
‘ObservationPoint’	140395.9	29	4841.237	7736.206	0.00
‘HoldoutSet’	33.7	7	4.814	7.693	0.00
‘Regressor ∗ ObservationPoint’	591604.8	116	5100.041	8149.770	0.00
‘Regressor ∗ HoldoutSet’	16.8	28	0.599	0.958	0.52
‘ObservationPoint ∗ HoldoutSet’	1297.4	203	6.391	10.213	0.00
‘Error’	508.1	812	0.626
‘Total’	1040349.2	1199