Population Nutrition-Driven Crop Recommendations: Addressing Regional Micronutrient Deficiencies for Improved Public Health

Crop yield optimization is paramount for an agriculture-based economy such as India, particularly in rural areas, which account for 92.02% of the agricultural workforce. However rural areas suffer the most from nutritional deficiencies. The proposed work attempts to combat the malnutrition by recommending crops which not only leverages the soil conditions (N, P, K, pH) and environmental conditions (humidity, rain, temperature) but also the micronutrient deficiencies among the regional population (Sodium, Potassium, Vitamin C, Calcium, Iron, Vitamin B6, Magnesium per 100mg). A hybrid dataset combining the crop recommendation dataset and USDA nutritional values dataset was prepared and utilized Six Machine Learning models-Decision trees, Support Vector machine (SVM), Logistic regression (LR), XGBoost, Random Forest (RF), and Naive Bayes (NB)-were trained and evaluated to determine the most effective model. The Random Forest classifier demonstrates an accuracy of 100<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> for the dataset used. The implementation of precision farming techniques optimizes crop yield production and contributes to food security and sustainable agriculture. Furthermore, the proposed work addresses the previously overlooked issue of deficiencies and recommends crops containing nutrients based on regional requirements. This approach will significantly impact agricultural produce and the holistic health of the population.