Open AccessRainfall Prediction for Udaipur, Rajasthan using Machine Learning Models Based on Temperature, Vapour Pressure and Relative Humidity
Author(s) -
Jitendra Shreemali,
Praveen Galav,
Gaurav Kumawat
Publication year - 2020
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.f1024.0386s20
Subject(s) - relative humidity , humidity , apparent temperature , environmental science , predictive modelling , vapor pressure , meteorology , mathematics , statistics , thermodynamics , geography , physics
The study aims at Rainfall prediction using Machine Learning models using the minimum of features. The prediction here is based on temperature, vapour pressure and relative humidity. Numerous studies carried out earlier used more features than this study. A training-test split of 75-25 was used. The best results were obtained by combining the best of the candidate models into an ensemble model to identify that predictor importance of vapour pressure was 0.89 while that of relative humidity was 0.11 with temperature not seen as a significant predictor for rainfall though the high correlation of temperature (°C) with vapour pressure (Torr) and relative humidity (Percentage) suggests that the two predictor variables subsume the impact of temperature.