Assessment for Thermal Conductivity of Frozen Soil Based on Nonlinear Regression and Support Vector Regression Methods

Table 4

Statistics of prediction results of the ternary fitting model and SVR model.

Predictive model	Statistical values	Silty clay	Silt	Weathered rock	Fine sand	Gravel sand	Boulder	Breccia
Ternary fitting method	R^2	0.76	0.79	0.9	0.91	0.93	0.87	0.78
	MAPE ≤ 10%	63.30%	68%	76.36%	76%	76.74%	77.08%	65.38%
		26.6%	22.88%	16.78%	19.62%	22.11%	19.73%	25.3%
SVR model	R^2	0.84	0.86	0.91	0.92	0.94	0.94	0.85
	MAPE ≤ 10%	74.74%	83.64%	72.55%	78.43%	79.07%	83.33%	69.81%
		20.47%	19.89%	19.62%	17.49%	15.62%	13.23%	20.04%