Enhanced Rainfall Prediction using Hybrid Machine and Deep Learning Models Across Diverse Climate Zones

1SIREESHA NEERUKOTA, 2L GANESH, 3KARE NEETHU LAKSHMI,4N VASANTH KUMAR,5
NARA AKHIL KUMAR
1Assistant Professor,Department of Information Technology, SV college of Engineering, Tirupati, India
2B.Tech, Department of Information Technology, SV college of Engineering, Tirupati, India
3B.Tech, Department of Information Technology, SV college of Engineering, Tirupati, India
4B.Tech, Department of Information Technology, SV college of Engineering, Tirupati, India
5B.Tech, Department of Information Technology, SV college of Engineering, Tirupati, India
Email: sireesha.n@svce.edu.in,lachannagariganeshyadhav@gamil.com,
lakshmineethu05@gmail.com,nandyalavasanthkumar143vs@gmail.com, , akhilnara83@gmail.com
Corresponding Author/Guide: Sireesha Neerukota, M.Tech(Ph.D), Assistant Professor

ABSTRACT:Accurate rainfall prediction is critical for various sectors including agriculture, disaster management, and resource planning.Accurate rainfall prediction uses advanced machine learning and deep learning models to analyze meteorological data and forecast precipitation patterns. It enhances weather forecasting accuracy by capturing complex, non-linear relationships among climatic factors. The existing system analyzes meteorological parameters using machine learning and deep learning models trained on five years of weather data from the United States, Canada, and Ireland. It applies correlation and feature importance analyses to identify key factors affecting rainfall prediction and evaluates models such as SVM, CART, 1D CNN, and LSTM. However, its limitations include limited geographical scope, inconsistent dataset feature depth, and difficulty modeling non-linear interactions across diverse climate zones. To address these limitations, the proposed system aims to extend the methodology to additional geographically and climatologically diverse locations, incorporating largerand higher-dimensional datasets. It will focus on extracting and utilizing a reduced subset of highly influential meteorological variables, thereby enhancing prediction accuracy while reducing computational complexity. The proposed system is expected to benefit from improved generalization across climate zones, more efficient data utilization, and stronger interpretability of rainfall prediction models, facilitating better-informed decision-making in climate-sensitive sectors. KEYWORDS: Meteorological data, Geographical Scope, Machine Learning, Generalization, Climate-sensitive sectors,Deep Learning.