Author(s): Chandra Sekhar Sanaboina

Abstract: The main objective of this paper is to find the best machine-learning technique for rainfall prediction. Predicting rainfall is a challenging and uncertain process that greatly impacts society. Timely and accurate forecasting can proactively reduce human and financial loss. In this article, a prediction model has been developed to forecast whether it will rain or not in major Australian cities based on the weather data on a given day. This data has been used to test the accuracy of several machine learning algorithms, such as logistic regression, random forest classifier, decision trees, neural networks, light GBM, CatBoost, and XGBoost. The results of this research article depicted that CatBoost, XGBoost, and LightGBM performed better when compared to the other models, with accuracy scores of 95%, 94%, and 94%, respectively. However, if speed is a priority, we can continue with LightGBM rather than XGBoost or CatBoost.

DOI: 10.33545/27076571.2024.v5.i2c.117

Pages: 211-219 | Views: 1391 | Downloads: 604

Download Full Article: Click Here