Treffer: Mapping landslide susceptibility in west bandung regency using artificial neural network (ANN) method.

Landslides are among the most destructive natural disasters, particularly in areas with steep topography. West Bandung Regency, located in West Java Province, is highly susceptible to landslides. This study applied the Artificial Neural Network (ANN), a machine learning algorithm, to map landslide susceptibility in the region. Eleven conditioning factors were used in the model, including elevation, slope angle, plan and profile curvatures, terrain ruggedness index (TRI), topographic position index (TPI), topographic wetness index (TWI), soil types, land use, NDVI, and rainfall. The ANN model was developed using Python, and its performance was evaluated through accuracy assessments using the Confusion Matrix (CF), ROC, and AUC scores. The model achieved an accuracy of 0.72 for the training-testing data and 0.55 for the validation data. ROC-AUC scores of 0.786 and 0.776 were obtained for training-testing and validation, respectively. Landslide susceptibility was classified into five categories: very low (28.88%), low (44.86%), moderate (7.52%), high (18.73%), and very high (11.88%). The findings demonstrate that the ANN model is an effective tool for mapping landslide susceptibility, providing critical insights for disaster risk management in the West Bandung Regency. This research contributes to improving disaster mitigation strategies and enhancing land use planning in vulnerable regions. [ABSTRACT FROM AUTHOR]

Copyright of AIP Conference Proceedings is the property of American Institute of Physics and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)