Rainfall Forecasting in India Using Combined Machine Learning Approach and Soft Computing Techniques : A HYBRID MODEL

Accurate rainfall prediction in India is crucial for agriculture, water management, and disaster preparedness, particularly due to the reliance on the southwest monsoon. This paper examines historical rainfall trends from 1901 to 2022, highlighting significant anomalies and changes identified through the Pettitt test. The effectiveness of advanced machine learning techniques is explored particularly the Artificial Neural Network-Multilayer Perceptron (ANN-MLP) in enhancing rainfall forecasting accuracy and compared with statistical methods. By integrating important climate variables—temperature, humidity, wind speed, and precipitation into the ANN-MLP model, its ability to capture complex nonlinear relationships is demonstrated. Additionally, the analysis employs geo-statistical techniques, specifically Kriging, to visualize spatial-temporal rainfall variability across different regions in India. The findings emphasize the potential of modern computational methods to overcome traditional forecasting challenges, ultimately improving decision-making for agricultural planning and resource management in the face of climate variability.