Author Name
Manvendra Singh, Dr. Sanjeev Kumar Sharma, Dr Subhash Mishra, Saurabh Karsoliya
Abstract
Accurate crop yield prediction is a critical component of precision agriculture, enabling efficient resource management, improved food security, and informed decision-making for farmers and policymakers. Increasing climate variability and heterogeneous agricultural conditions have reduced the effectiveness of traditional statistical yield estimation techniques. To address these challenges, this paper presents a robust and scalable machine learning framework for crop yield prediction using advanced ensemble learning methods. The proposed study evaluates the performance of Random Forest, XGBoost, and Light Gradient Boosting Machine (LightGBM) on a large-scale agricultural dataset comprising climatic, soil, crop, and management-related attributes. A unified preprocessing and modeling pipeline is employed to ensure fairness and reproducibility.