Result: Simulating the rate of prediction error in rainfall big data forecasting.

Farmers can maximize their productivity by forecasting rain quantity accurately, and people can access food and water that they need to stay healthy by having access to accurate rain forecasts. Rainfall forecasting has been the subject of numerous studies, many of which employ data mining and machine learning methods applied to environmental datasets from a variety of nations. The agriculture industry, which drives the economy, suffers when rain falls in unpredictable patterns. The problems with drought and flooding can be mitigated by better planning and implementation of the use of rainwater. In this paper, we intend to model several machine learning supervised algorithms to forecast the rainfall from the collected datasets to test the prediction error during forecasting. Various machine learning models are modelled using python simulator to test the efficacy of the model in forecasting and the rate at which the error occurs during the process of forecasting. In this study, we use several supervised machine learning algorithms that gets trained and tested over severalcollected rainfall big datasets. The dataset is collected from several repositories that are of various categories, where performance metrics such as accuracy, precision, recall, and f-measure are used to measure the model's performance. [ABSTRACT FROM AUTHOR]