Paper Key : IRJ************398
Author: K. Nikhil Varma,Mrs. K Ashwini
Date Published: 02 Mar 2025
Abstract
Rainfall estimation is crucial for various applications, including agriculture, hydrology, and climate studies. Traditional methods rely on sparse ground-based measurements or satellite data, which may lack accuracy in specific regions. This paper presents a machine learning (ML) approach that integrates heterogeneous data sources, including satellite imagery, weather station data, and remote sensing information, to improve rainfall estimation accuracy. We explore multiple ML models, such as Random Forest (RF), Support Vector Machines (SVM), and deep learning architectures, to analyze spatiotemporal patterns in rainfall prediction. Our results demonstrate the efficacy of integrating diverse datasets and employing ML techniques in achieving more precise rainfall estimations.