Machine Learning Approaches for Rainfall Trend Analysis: Insights from Precipitation and Meteorological Data

Rainfall significantly impacts agriculture, water resources, and natural disasters like floods and droughts. Understanding rainfall trends is crucial for effective planning and mitigation. This study analyses rainfall trends using machine learning models trained on precipitation data and four meteorological features: temperature, specific humidity, relative humidity, and wind speed. Five models Linear Regression, KNN, SVR, Random Forest, and Gradient Boosting were evaluated using k-fold cross-validation and performance metrics, including Mean Squared Error (MSE) and R² Score. Among these, Random Forest outperformed the others with the lowest MSE (15.79) and the highest R² Score (71.69%), demonstrating its ability to capture seasonal trends. Gradient Boosting followed closely with an R² Score of 68.68%, while KNN achieved a moderate prediction accuracy with an R² Score of 67.97%. These findings highlight the potential of machine learning models for rainfall prediction, offering valuable insights for water resource management, disaster preparedness, and agricultural planning.