Table 9.

Results of ensembles landslide modeling using combination of 90%/10% and raster resolution of 20 m.

Criteria	ADTree		RF		RS		BA		MB
	T	V	T	V	T	V	T	V	T	V
True positive	89	10	96	10	88	10	87	10	92	10
True negative	92	10	94	10	92	9	93	10	95	10
False positive	11	1	4	1	12	1	13	1	8	1
False negative	8	1	6	1	8	2	7	1	5	1
Sensitivity	0.918	0.909	0.941	0.909	0.917	0.833	0.926	0.909	0.948	0.909
Specificity	0.893	0.909	0.959	0.909	0.885	0.900	0.877	0.909	0.922	0.909
Accuracy	0.905	0.909	0.950	0.909	0.900	0.864	0.900	0.909	0.935	0.909
AUROC	0.957	0.876	0.983	0.913	0.968	0.884	0.968	0.921	0.992	0.926

T, training dataset; V, validation dataset.