DETECTION OF LANDSLIDE USING MACHINE LEARNING

Abstract

Landslides occur when large amounts of earth, rock, sand or mud flows swiftly down hill and mountain slopes.This project proposes a novel machine-learning method to identify the reason of landslides using global landslide dataset.The five machine learning algorithm, including Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), Boosting Method and Decision Tree are utilized and evaluated on global landslide dataset.An Ensemble Method is also used to evaluate the global landslide dataset.From the result, Ensemble method gives an accuracy of 90% . Random Forest(RF) comes in second with 89% accuracy.By using machine learning

technique ,the proposed landslide reason identification shows outstanding robustness and great po- tential in tackling the landslide reason identification problem.