XGBoost-Based Supervised Learning Approach for Crop Selection in Jharkhand

This is the authors' abstract. We don't add key points for this paper.

Agriculture is the fundamental of India’s economy, and using the concept of Data Driven Farming presenting an important opportunity to boost national productivity. In Jharkhand more than 80% of the rural population relies on agriculture yet it remains as one of India's poorest state(as of 2020). The region’s observance to traditional and non-data-driven farming methods has resulted in inactive crop yields and hindered economic progress. In this Paper a method is proposed based on XGBoost-Based Supervised Learning. The proposed model is trained on identical historical data, including temperature, rainfall, irrigation, and soil nutrient composition which includes Nitrogen (N), Phosphorus (P), and Potassium (K). The study found that the proposed method delivered the most accurate crop yield predictions, identifying it as the optimal tool for modernizing and improving agricultural outcomes in Jharkhand.