Yapay Sinir Ağları ile Heyelan Duyarlılık Haritalarının Üretilmesinde Farklı Algoritmaların Kullanımının Araştırılması

Abstract

This study aims to investigate the effects and performances of the uses of different learning algorithms in the training of artificial neural networks on the resulting maps by applying the artificial neural networks (ANN) method, which is commonly used for the creation of landslide susceptibility maps in the literature. For this purpose, a study area is selected within the borders of Karabük province in the Western Black Sea Region. By utilizing site investigations, aerial photographs of the study area and Google Earth satellite images, 196 landslides were mapped and a landslide database was created. In order to be used in the analysis of landslide susceptibility: topographical elevation, slope angle, aspect, wetness index, lithology and vegetation index parameters were taken into account.