A Comparative Assessment of Sampling Ratios Using Artificial Neural Network (ANN) for Landslide Predictive Model in Langat River Basin, Selangor, Malaysia

Landslides have been classified as the most dangerous threat around world, causing huge damage to properties and loss of life. Increased human activity in landslide-prone areas has a major contributor risk landslide occurrences. Therefore, machine learning used studies develop predictive model. The main objective this study is evaluate suitable sampling ratio for model Langat River Basin (LRB) using Artificial Neural Networks (ANNs). inventory was divided randomly into training testing datasets four ratios (50:50, 60:40, 70:30, 80:20). A total 12 conditioning factors were considered study, including elevation, slope, aspect, curvature, topography wetness index (TWI), distance road, river, faults, soil, lithology, land use, rainfall. evaluation performed certain statistical measures area under curve (AUC). Finally, chosen based on validation results compound factor (CF) method. Based results, with an 80:20 indicates realistic finding first rank among others. AUC value dataset 0.931, while 0.964. These attempts will help great deal when it comes choosing best samples create reliable complete prediction LRB.