Evidential reasoning applied to GIS-based landslide susceptibility mapping with geospatial data

This paper applies an evidential reasoning model to integrate multiple geospatial data for landslide susceptibility mapping. A data-driven information representation procedure based on likelihood ration functions is applied to assign mass functions. After defining the mass functions is applied to obtain a series of combined mass functions. A case study in the Jangheung area in Korea was carried out to illustrate the applicability of this methodology. The results of the case study showed that the the presented methodology efficiently represented and integrated multiple geospatial data and showed better prediction capability than that of a logistic regression model.