Authors: Harshal Patel, Jitendra Shrivastav, Kamlesh Patidar
Abstract: The integration of machine learning into agriculture has shown great promise in improving decision-making, particularly in crop recommendation systems. This review focuses on XGBoost and supervised learning approaches for crop recommendation based on soil composition analysis. Soil properties such as pH, nutrient levels, moisture, and texture play a crucial role in determining crop suitability and yield. However, many existing models fail to adequately capture the complex interactions among these variables or account for regional soil variability. By examining current supervised learning methods and the growing application of XGBoost, this paper highlights their strengths, limitations, and potential for enhancing prediction accuracy. Furthermore, it identifies key research gaps, including the scarcity of diverse soil–crop datasets and the need for models that can adapt across geographical regions and climates. The review concludes that integrating advanced supervised learning with robust soil data can significantly optimize crop recommendations, promoting sustainable and precise agricultural practices.
