Hybridization of Machine Learning Algorithms and an Empirical Regression Model for Predicting Debris-Flow-Endangered Areas

Table 1

Ranges of V_D, H, and L_f at training and testing stages.

Parameters	VD (×10^3 m^3)	H (km)	Lf (km)
Total data	1.2–3238.4	0.22–2.98	0.03–0.88
Cross-validation	Training
CR1	2.7–3238.4	0.22–2.98	0.03–0.88
CR2	1.2–3238.4	0.22–2.48	0.03–0.88
CR3	1.2–3238.4	0.22–2.98	0.03–0.88
CR4	1.2–3238.4	0.26–2.98	0.03–0.88
CR5	1.2–3100.0	0.22–2.98	0.038–0.51
Cross-validation	Testing
CR1	1.2–711.0	0.34–2.48	0.04–0.39
CR2	3.0–561.0	0.36–2.98	0.042–0.51
CR3	2.9–1034.0	0.26–1.60	0.038–0.50
CR4	4.5–3100.0	0.22–2.02	0.05–0.49
CR5	2.7–3238.4	0.26–1.98	0.03–0.88