Environmental Management Hazards 2423-415X 2 2 2015 06 22 Comparison of Artificial Neural Network Model With Analytical Hierarchy Process In Landslide Hazard Assessment Using Geographic Information Systems Comparison of Artificial Neural Network Model With Analytical Hierarchy Process In Landslide Hazard Assessment Using Geographic Information Systems 225 250 55063 10.22059/jhsci.2015.55063 FA Imanali Balvasi M.Sc. Graduated of Remote Sensing & Geographic Information Systems, University of Tabriz, Tabriz, Iran Mohammad Hossein Rezaei Moghaddam professor of Geomorphology, Faculty of Geography and Planning, University of Tabriz. Mohammad Reza Nikjo Assistance Professor, Department of geomorphology, Faculty of Geography & Planning, University of Tabriz, Tabriz, Iran. Khalil Valizadeh Kamran Assistance Professor, Department of Climatology, Faculty of Geography & Planning, University of Tabriz, Tabriz, Iran. Journal Article 2015 02 20 Landslide is one of the natural hazards in mountainous regions that results in huge losses every year. Alashtar Doab watershed with mountainous terrains, uplands and different natural conditions has the potential for landslide. The purpose of this study is to compare the ANN^{^[1]} model with AHP to evaluate landslide in Alashtar Doab watershed. In order to preparing the map, first of all parameters of the landslide were extracted and then the layers were prepared and after that a landslide distribution map that was occurred in the basin was prepared and then by combining landslide influencing factors with landslide distribution map, the impact of each of these factors such as slope, aspect, elevation, lithology, rainfall, land use, distance from fault and stream in ArcGIS software were measured. In this study, in order to landslide hazard zoning in Alashtar Doab watershed, the ANN and AHP^{^[2]} were used. Back propagation algorithm and sigmoid activation function were used in ANN. The final structure of the network consisted of eight neurons in the input layer, eleven neurons in the hidden layer and one neuron in the output layer. After optimization of the network structure, all area information was imported to the network and finally, landslide hazard zoning map was prepared according to output weight. In AHP method, after paired comparisons and extracting of the weight of parameters, the potential landslide area was obtained by combining them. The kappa statistic factor was used for assessment and classification output results of model that were used to estimate of landslide hazard. The result shows that the ANN model with 0.9 kappa coefficient is more efficient Landslide is one of the natural hazards in mountainous regions that results in huge losses every year. Alashtar Doab watershed with mountainous terrains, uplands and different natural conditions has the potential for landslide. The purpose of this study is to compare the ANN^{^[1]} model with AHP to evaluate landslide in Alashtar Doab watershed. In order to preparing the map, first of all parameters of the landslide were extracted and then the layers were prepared and after that a landslide distribution map that was occurred in the basin was prepared and then by combining landslide influencing factors with landslide distribution map, the impact of each of these factors such as slope, aspect, elevation, lithology, rainfall, land use, distance from fault and stream in ArcGIS software were measured. In this study, in order to landslide hazard zoning in Alashtar Doab watershed, the ANN and AHP^{^[2]} were used. Back propagation algorithm and sigmoid activation function were used in ANN. The final structure of the network consisted of eight neurons in the input layer, eleven neurons in the hidden layer and one neuron in the output layer. After optimization of the network structure, all area information was imported to the network and finally, landslide hazard zoning map was prepared according to output weight. In AHP method, after paired comparisons and extracting of the weight of parameters, the potential landslide area was obtained by combining them. The kappa statistic factor was used for assessment and classification output results of model that were used to estimate of landslide hazard. The result shows that the ANN model with 0.9 kappa coefficient is more efficient https://jhsci.ut.ac.ir/article_55063_ae89f84e1461a21d5a22d82c4db17851.pdf