Prediction of California Bearing Ratio of Granular Soil by Multivariate Regression and Gene Expression Programming

Table 10

Performances of multivariate and GEP models.

Model	MAE	MSE	RMSE	MAPE	R ^2	U95	Reliability
MLR1	18.177	563.803	23.745	24.28	0.75	15.940	49
MLR2	16.546	432.261	20.791	24.51	0.81	15.569	51
MLR3	19.208	593.443	24.361	28.12	0.74	16.033	49
MLR4	24.952	890.208	29.836	41.58	0.61	16.833	43
MLR5	17.783	474.159	21.775	27.34	0.79	15.699	49
MLR6	17.618	513.078	22.651	26.83	0.78	15.816	49
MLR7	16.698	478.326	21.871	24.69	0.79	15.702	53
PQ1	15.933	485.939	22.044	18.94	0.79	15.723	55
PQ2	11.560	244.783	15.646	16.75	0.89	15.022	62
PQ3	15.822	470.641	21.694	21.35	0.79	15.678	57
PQ4	16.843	470.530	21.692	26.93	0.79	15.678	51
PQ5	11.490	256.206	16.006	16.17	0.89	15.084	57
PQ6	11.744	293.372	17.128	17.16	0.87	15.174	57
PQ7	12.087	273.340	16.533	18.00	0.88	15.127	62
GEP1	18.251	509.44	22.552	24.79	0.776	15.787	53
GEP2	16.401	523.861	22.888	20.31	0.771	15.828	53
GEP3	16.878	532.636	23.0789	21.195	0.766	16.803	45
GEP4	18.471	510.715	22.599	28.393	0.776	16.766	45