Landslides are one form of geohazard that is crucial for risk assessment and mitigation. This study proposes machine learning methods to identify landslides using data collected from Muzaffarabad, Pakistan. We applied classification techniques including Random Forest, Decision Tree, Gradient Boosting Machine (GBM), light gradient boosting machine (LGBM), neural network (NN), 1D-CNN (convolutional neural network one dimensional), XGBoost, and Catboost. Our dataset consists of 12 different factors and data of 1213 sites to predict a landslide. Initially, we trained models individually and compared their performances to find a model with the highest accuracy. In the end, multiple performance measures are used for the evaluation of applied models. From the results, we achieved the highest test accuracy of 84% using NN and boosting methods, with more than 80% accuracy. Our proposed system can be effectively used to predict landslides for other sensitive regions with comparable environmental conditions.