• Jaishree Dubey
• Swapnil Shinde
• Janhavi Bilgaye

Forecasting rainfall is essential for agricultural planning, disaster preparedness, and sustainable water resource management. Traditional physics-based models often struggle with region-specific variability and computational complexity. In recent years, machine learning has emerged as a viable alternative due to its ability to capture non-linear, temporal, and spatial dependencies in climatic data [1][2]. This paper presents a district- level rainfall prediction system for India, built using a Random Forest Regressor trained on historical rainfall data from 2017 to 2023. The model is designed to forecast monthly rainfall from April 2025 to December 2026 with high accuracy. Random Forest was selected for its proven robustness, interpretability, and resistance to overfitting in regression problems involving environmental data [3]. The system is deployed using a lightweight Flask web framework that enables users to interactively select their state and district, generate month-wise predictions, and visualize results through dynamic graph generation using matplotlib. The final output includes both tabular and graphical formats, which enhance interpretability and usability. The proposed system offers a practical, scalable, and user-friendly solution for stakeholders such as farmers, researchers, and disaster planning authorities who require timely, localized rainfall insights.

- Rainfall Forecasting, Machine Learning, Random Forest, Flask Web App, Weather Prediction, Graph Generation, Result Optimization

"Rainfall Prediction Using Machine Learning: A Flask-Based Regional Forecasting System for Indian Districts", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 4, page no.l315-l326, April-2025, Available :http://www.jetir.org/papers/JETIR2504B36.pdf

"Rainfall Prediction Using Machine Learning: A Flask-Based Regional Forecasting System for Indian Districts", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 4, page no. ppl315-l326, April-2025, Available at : http://www.jetir.org/papers/JETIR2504B36.pdf