A Machine Learning Model for Estimation of Village Level Soil Nutrient Index

Objectives: To propose an innovative technique for designing efficient and adaptive machine learning model using classifier assembly estimating village level soil nutrient index datasets. Methods: Freely available datasets were collected from the concerned authority of Govt. India. These used by proposed designed with a fifteen diverse classifiers class identification. The performance each was evaluated in terms five well-accepted standard metrics. outputs best performing then estimation modified Parker’s method. Findings: applied identification, different villages freely benchmarked Soil health Card provided empirical results depicted that this overperformed other existing models average accuracy score. In case Copper, it highest classification (0.949) 95.48%. For Sulphur, 0.891 90.66% achieved. Similarly, Zinc, 0.883 89.63% observed. Novelty: This study suggests novel architecture to estimate possible accuracy, Keywords: Nutrient index; Village fertility; Fertilizer management; Machine learning; Classifier