Landslide Susceptibility Mapping Using Classification Rule Discovery by Ant Colony Optimization and GIS

Landslide susceptibility mapping is a fundamental tool for disaster management. The purpose of the present study is to investigate the landslide susceptibility mapping using Classification Rule Discovery (CRD) by Ant Colony Optimization (ACO). This modeling approach was applied for Landslide susceptibility assessment in Javanroud county of Kermanshah province.For this purpose, thematic layers including slope, distance to faults, distance to stream, rainfall, land use, and soil texture were used. The One- At-a-Time (OAT) approach was utilized as the sensitivity analysis method to determine the dependency of model outcomes on the input parameters. Then the performance of the proposed algorithm was validated by comparing it with C5 decision tree algorithm, which is a well-known classification rule discovery method. To assess the accuracy of the resulting landslide susceptibility map, it was evaluated by the distribution of the observed landslides. The resulting map showed that the predictive power of the model is very high. Overall, about 20% of the study area falls in susceptible and very susceptible classes, and most of the previous landslides (81.25%) occur in the same classes.The results also indicated that the model can be effectively used in preparation of landslide susceptibility maps.