Improving the Landslide Susceptibility Prediction Accuracy by Using Genetic Algorithm Optimized Machine Learning Approach

Table 6

Summary of results of binary logit regression analysis.

Items	Regression coefficient	Standard error	z	Wald χ^2		OR	OR 95% CI
Slope height	0.002	0.001	1.448	2.098	0.148	1.002	0.999 ∼ 1.005
Slope angle	−0.045	0.017	−2.607	6.797	0.009	0.956	0.924 ∼ 0.989
Unit weight	0.202	0.053	3.849	14.814	0.001	1.224	1.104 ∼ 1.357
Cohesion	0.002	0.005	0.529	0.279	0.597	1.002	0.994 ∼ 1.011
Friction angle	0.086	0.023	3.806	14.487	0.001	1.090	1.042 ∼ 1.139
Pore water pressure	0.370	0.833	0.444	0.197	0.657	1.448	0.283 ∼ 7.406
Intercept distance	−5.738	1.033	−5.557	30.877	0.001	0.003	0.001 ∼ 0.024

Dependent variable: stability. McFadden R²: 0.225. Cox and Snell R²: 0.259. Nagelkerke R²: 0.352.