Abstract
Abstract. In this paper, Radial Basis Function (RBF) Neural Network and Logistic Regression (LR) models were proposed for hazard prediction of landslides in a part of the Semirom area (Iran) to compare their accuracy and performance. For this purpose, a spatial database of the study area was prepared that consists of 68 landslide locations and 11 influencing information layers including slope, aspect, profile curvature, plan curvature, distance from faults, distance from roads, distance from residential regions, distance from rivers, land use, lithology and rainfall. Landslide hazard maps were prepared for the study area by applying the proposed algorithms. Performance of the models was assessed using the Receiver Operating Characteristic (ROC) curve and area under the ROC curve (AUC). The coefficient of determination (R2), the root mean square error (RMSE), and the Normal Root Mean Square Error (NRMSE) were calculated for proposed methods. The outcomes showed that the RBF Neural Network has the highest R2 (0.8224), in comparison to that of the LR model (0.5365). Also, the ROC plots, RMSEs and NRMSEs showed that the proposed RBF Neural Network is much better than the LR model. Consequently, it can be concluded that the RBF Neural Network is the best regression model in this study and it can be considered as a capable method for landslide hazard mapping in landslide-susceptible areas.
Highlights
One of the most sensitive and vital issues in the development projects, such as selecting the highway route and any development of mining and construction, depends on studying the sustainability of the land of the desired project
Determining landslide-susceptible areas is an important phase in land use planning
An appropriate land use plan can help managers to decrease financial injuries and loss of life resulted by landslides
Summary
One of the most sensitive and vital issues in the development projects, such as selecting the highway route and any development of mining and construction, depends on studying the sustainability of the land of the desired project. Geospatial information system and geostatistical analyses have been widely applied in analyzing different field of georelated topics (Delavar, 2004; Pahlavani et al, 2006; Pahlavani et al, 2017; Bahari et al, 2014). In this regard, applying statistical approaches to forecast landslide hazard has generally been seen since the 1990s. The approaches that have been used to generate a landslide hazard zonation map divided into four classes (Youssef et al, 2016): Heuristic, Deterministic, Statistical and combination of the cited approaches. Some researchers classify the landslides effective layers into two classes of internal and external
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
