Precision Rainfall Predictions: A Daily Weather Data Approach using Machine Learning

Abstract

Rainfall prediction is an important task since a lot of individuals rely on it, especially in agriculture. The study attempts to predict rainfall using machine learning algorithms, taking into account the impact of shortages or excessive rainfall on rural and urban life. Several techniques and approaches for predicting rain have been developed; however, there is still a lack of precise outcomes. The comparative study focused on incorporating Machine Learning (ML) models, analyzing various situations and time horizons, and predicting rainfall by using three different approaches. This research uses data preprocessing, feature selection, and machine learning methods like Random Forest, K-nearest neighbor (KNN), and Logistic Regression. This study shows the usefulness of machine-learning approaches in forecasting rainfall. In comparison, Random Forest performs better when compared to other models with a high precision rate.