A REVIEW PAPER ON MACHINE LEARNING BASED MODELS FOR RAINFALL PATTERN PREDICTION

Abstract

Rainfall is a pivotal rainfall parameter in the environment of India. vaticination of downfall can effectively prop the decision-making process for husbandry and natural disaster operation of the country. still the chaotic nature of downfall due to climate change has made the task of downfall vaticination challenging through traditional statistical models. In this study, we dissect the performance of machine literacy algorithms Decision Tree (DT), K- Nearest Neighbours (KNN), Random Forest (RF), Extreme Gradient Boosting (XGB), Light grade Boosting (LGB) and Multi-Layered Perceptron (MLP) in prognosticating diurnal downfall as both retrogression and bracket. The models were trained with and without point selection and/ or slice ways (for bracket). During training10-fold cross confirmation and hyper parameter tuning was performed on the train set and latterly the named models were applied to the test set for evaluation.

Keywords