Floods and Landslides are among the most devastating natural disasters, causing significant loss of life, damage to infrastructure, and disruption of livelihoods. With climate change, rapid urbanization, and environmental degradation, the frequency and intensity of these disasters have increased globally. Effective prediction and early warning systems are critical in mitigating their impacts and improving disaster preparedness. This work proposes a Machine Learning- based approach for flood and landslide prediction by analyzing environmental factors such as monsoon intensity, deforestation levels, urbanization, topographical changes, and climatic variations. The dataset utilized for this work includes multiple influencing parameters. Various Machine Learning models, including Logistic Regression, Decision Trees, Random Forests, Support Vector Machines (SVM), K-Nearest Neighbours (KNN), Naïve Bayes, Ridge Regression, XGBoost, and Gradient Boosting, are employed to predict flood and landslide probabilities. The predictive Modeling process begins with feature selection, where significant environmental variables contributing to floods and landslides are identified. Data preprocessing techniques such as Normalization and Standardization are applied to improve model efficiency. Several performance metrics, including Accuracy, Precision, Recall, F1-score, and RMSE, are used to assess model effectiveness. Results from the study indicate that Logistic Regression performs best in classifying flood-prone areas, achieving a good accuracy. Similarly, Ridge Regression and Gradient Boosting models are effective in estimating the severity of landslides.
Keywords : SVM, K- Nearest Neighbors (KNN), Naïve Bayes, Ridge Regression, XGBoost, and Gradient Boosting
Author : P GangadharaReddy , J Pradeep , P Naga Subba Rayudu , T Ramashri
Title : Flood and Landslide Prediction Using AI and Ensemble Machine Learning Models
Volume/Issue : 2025;2(2 ( April - June ))
Page No : 54 - 59