Comparison of Time-Series Forecasting Models based on Prophets for Predicting Rainfall

Forecasting rain is crucial from both the perspective of society and the economy. The preparation process of farming, drinking water systems, and other macroeconomic proiects depends on accurate rainfall forecasts. The amount of precipitation is a crucial climatic component and valuable environmental asset that has a wide range of effects on farming and daily life. For humans to produce and live, precisely correct weather predictions are crucial. The first goal of this research endeavor is to determine if Prophet-based systems are capable of forecasting rainfall across time series. In this study, time-series forecasting models based on prophets were compared for predicting rainfall. In terms of Root Mean Square Error (RMSE), it was discovered that Neural Prophet (NP) outperformed Facebook Prophet (FBP). When the actual and anticipated values of both Neural Prophet and Facebook Prophet were compared, it was found that Neural Prophet's predictions were more accurate than those of Facebook Prophet. It is observed from the experimental analysis that the RMSE of FBP is 2.85 and the RMSE of Neural Prophet is 2.25.