A Comprehensive Assessment of Machine Learning Techniques for Temperature and Rainfall Forecasting: SVR vs. Decision
K. Rajathi, Alex David S, Ruth Naveena N, Ravi Kumar Suggala, Almas Begum, D Hemalatha
Abstract
Machine Learning models have been used to predict the Monthly Rainfall and Night mean Temperatures of Chennai, which allows us to find out best model to predict them in respective assignment. The primary objective is to develop accurate and reliable prediction models to aid in better environmental and agricultural planning. The research employs Support Vector Regression (SVR) and Decision Tree Regressor models, following a robust preprocessing pipeline to address missing data and select significant features critical for prediction accuracy. We evaluate the models using performance metrics like Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), R-squared (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>), Mean Squared error(MSE) and Explained variance score(EVS). Results demonstrate that SVR excels in predicting night time mean temperatures, achieving RMSE, MAE, and R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> values of 0.9515, 0.6983, and 0.8693, respectively, showcasing high accuracy and low error. Conversely, the Decision Tree Regressor proves more effective for forecasting monthly rainfall, with an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.1465, MAE of 298.69, and RMSE of 399.62, despite the intrinsic challenges of rainfall prediction. The findings highlight that SVR is better suited for temperature prediction, while the Decision Tree Regressor offers a viable approach for rainfall forecasting. This study underscores the importance of selecting appropriate models based on the target variable and presents a framework for future research on environmental data modeling.