Assessment of logistic and multivariate regression Models for Landslide hazard zonation (Case Study: Marbor basin)

The aim of this research is Identification of Effective factors in Landslide Occurrence and its ýsensitivity Zonation Using logistic regression and multivariate linear regression. For this purpose, ýwith use of interpretation of arial photos with scale of 1:40000, geological, topographic maps ýand field survey using GPS, Landslide hazard map was prepared as dependent variables. For ýdetermination of effective factors in Landslide Occurrence, Analysis of numerical values of the ýparameters by using Support Vector Machines in Rapid Miner Software have been used and ýfrom 21 selective data layer, 15 data layer selected and for preparation of zonation map as the ýindependent variable in ArcGIS 10.1ýþ þPrepare and were digitized. After weighing the layers, ýZontion map were prepared using selective methodf in five classes: very low, low, moderate, ýhigh and very high. Result of weighting layers showed that in both methods, landuse and aspect ýhave the greatest impact on landslides. The ROC (Receiver operating characteristic) curves and ýarea under the curves (AUC) for landslide susceptibility maps were constructed and the areas ýunder curves were assessed for validation purpose and its values showed that multivariate linear ýregression model (0.890) has a higher efficiency than the logistic model (0.829) for landslide ýhazard zonation. ý