Crop Yield Estimation Using AdaBoost Regression Under Multivariate Environmental Conditions

Abstract

Accurate crop yield prediction is a critical component of agricultural planning, food security, and sustainable farming practices. However, crop yield is influenced by complex and nonlinear interactions among environmental factors and agricultural management practices, such as climatic conditions and pesticide usage, which are often inadequately modeled by traditional statistical approaches. Motivated by these limitations, this study proposes a crop yield prediction model based on AdaBoost regression that integrates multivariate climatic variables and pesticide usage data.The proposed approach employs an ensemble learning strategy to improve prediction accuracy by adaptively combining multiple weak regressors, enabling the model to capture nonlinear relationships and handle data variability effectively. A structured methodology is applied, including data preprocessing, feature normalization, model training, and hyperparameter tuning. The performance of the proposed model is evaluated using standard regression metrics, namely R-squared (R²), Mean Absolute Error (MAE), and Root Mean Square Error (RMSE).Experimental results demonstrate that the AdaBoost regression model achieves accurate and reliable crop yield predictions, with predicted values closely aligning with actual observations. The findings indicate that integrating climatic factors and pesticide usage within a single model provides a more comprehensive and realistic representation of agricultural systems. The main contribution of this research lies in the application of AdaBoost regression for crop yield estimation under multivariate environmental conditions, highlighting its robustness and suitability for precision agriculture. Future work may extend this framework by incorporating additional agronomic variables, such as soil properties and irrigation practices, and by exploring hybrid or deep learning-based approaches to further enhance prediction performance.