Landslides threaten not only human life but also infrastructure, calling for precise and trusted predictive models. This paper delves into landslide susceptibility mapping effectiveness using ensemble learning methods on the basis of Random Forest (RF), XGBoost, AdaBoost, and Stacking Classifier machine learning algorithms. The study uses a data set involving topographical, geological, hydrological, and meteorological attributes in training and model validation. Performance is measured based on metrics like accuracy, precision, recall, F1-score, and Area Under the Curve (AUC). The findings prove that ensemble learning has a profound improvement in the accuracy of predictions over single classifiers, with the Stacking model yielding the best performance. The research identifies the promise of ensemble learning in constructing reliable and interpretable landslide prediction systems for disaster risk reduction.