Rainfall Prediction using Random Forest and XGBoost-A comparative study

There are numerous aspects of human life where knowing how much rain to expect might be beneficial. Heavy rainfall events such as flash floods and landslides, as well as droughts, can be predicted with effective rainfall forecasting. Because of reliable weather forecasts, the infrastructure required to capture rainwater and cultivate crops may be planned ahead of time. With the advent of big data, new possibilities have arisen for weather prediction and research. A variety of Machine Learning and Deep Learning algorithms enable accurate weather forecasting. This study aims to investigate the efficacy of Random Forest in forecasting rainfall in Jammu Region of Union Territory of Jammu and Kashmir and compare its performance with XGBoost using RMSE as performance metric. Hyperparameter Tuning is used to enhance the efficiency of the algorithms. The dataset is collected for a period of 10 years from 2012 to 2022. Random Forest comes out to be slightly better than XGBoost when compared on test data by having less RMSE.