Abstract
A comparative study is done in this paper in the prediction of rainfall at ground level multiple linear regression and, feature selection and k-means clustering method. Based on the past observations of the last three days atmospheric parameters like minimum and maximum Temperature, minimum and maximum relative humidity, minimum and maximum air pressure, minimum and maximum vapour pressure and minimum and maximum radiation the model is developed. In this paper it is observed that considering the seasonality effect better results can be achieved. It has also been observed that the selection of appropriate features can also improve the performance of the prediction.
Recommended Citation
Mukherjee, Writaparna; Ray, Angshuman; Datta, Bimal; and Pal, Pintu
(2024)
"Prediction of Rainfall based on Statistical and Computational Approach,"
American Journal of Science & Engineering (AJSE): Vol. 3:
Iss.
2, Article 5.
Available at:
https://research.smartsociety.org/ajse/vol3/iss2/5