Landslip Trigger Classification of Global Landslide Catalog Dataset Using Machine Learning Algorithm

A severe natural calamity that results in loss of life and property is a landslip. As a result of rising urbanization, destruction of forests, and global warming, it is expected to happen more often than previously. Mapping the likelihood of landslides is essential for protecting people and property. Many research papers been conducted in the past to approximate the likelihood of landslides occurring based on terrain slope and soil type and even more recent studies have used machine learning techniques to predict sensitivity to landslides. The prediction of landslip triggers hasn't yet been studied, though. By identifying the main sources of landslides in various parts of the world, our research aims to advance the current body of knowledge. By identifying the main causes of landslides in different regions of the world, our research aims to advance the current body of knowledge. This paper summarizes five implementation techniques of machine learning: K-nearest neighbor, Support Vector Machine random forest, and Xgboost using a dataset of landslip incidents and their related triggers. The efficiency of all well-known ML classifiers was demonstrated. Ensemble tree techniques (i.e., Xgboost) have been found to have demonstrated to have been most successful in terms of overall performance.