Explainable AI for Soil Fertility Prediction
Harshiv Chandra, Pranav M. Pawar, R Elakkiya, P. S. Tamizharasan, Raja Muthalagu, Alavikunhu Panthakkan
Abstract
Soil fertility refers to the ability of soil in a particular area to provide favorable chemical, physical and biological characteristics that help the plant in its growth. It is affected by multiple parameters, from the available concentration of Nitrogen in the soil to the concentration of Organic Carbon in the soil. This paper discusses the implementation of an explainable AI (XAI) model based on a Random Forest classifier. The developed model reliably predicts the relative soil fertility of a given soil using its various physiochemical properties, and explain the reasons behind the model’s soil fertility indicator prediction using user friendly graphs. The model shows 97.02% accuracy in comparison with state-of-the-art machine learning models. The paper also discusses applications of developed model in providing possible solutions to further improve upon soil fertility in the short term and long term.