Application of artificial neural networks for rice and wheat crop yield prediction in Chhattisgarh
Author(s): Purnima Sahu, Sweta Ramole and Tripti Swarnkar
Abstract: Climate variability and increasing demand for food necessitate accurate crop yield prediction models to enhance agricultural planning and ensure food security. This study evaluates the predictive performance of three modeling techniques: Artificial Neural Network (ANN), Multiple Linear Regression (MLR), and Ridge Regression for rice and wheat yield forecasting in the Chhattisgarh Plains. Using 27 years (1997-2023) of climatic and yield data, including rainfall, temperature, humidity, and wind parameters, the models were trained and validated on standardized datasets. Results indicate that ANN outperformed other methods, achieving the highest accuracy (R² = 0.87 for rice and 0.94 for wheat) with the lowest error rates. Ridge Regression provided moderate accuracy, while MLR recorded the weakest performance with low explanatory power and higher errors. The findings underscore ANN’s effectiveness in capturing complex non-linear relationships between climatic variables and crop yields, making it a promising tool for precision agriculture and climate-adaptive farming strategies.
Purnima Sahu, Sweta Ramole, Tripti Swarnkar. Application of artificial neural networks for rice and wheat crop yield prediction in Chhattisgarh. Int J Stat Appl Math 2025;10(9):49-55. DOI: 10.22271/maths.2025.v10.i9a.2162
