Assessment of Soil Suitability Using Machine Learning in Arid and Semi-Arid Regions

Increasing agricultural production is a major concern that aims to increase income, reduce hunger, and improve other measures of well-being. Recently, the prediction soil-suitability has become primary topic rising among academics, policymakers, socio-economic analysts assess dynamics production. This work use physico-chemical remotely sensed phenological parameters produce maps (SSM) based on Machine Learning (ML) Algorithms in semi-arid arid region. Towards this goal an inventory 238 suitability points been carried out addition to14 4 have used as inputs machine-learning approaches which are five MLA prediction, namely RF, XgbTree, ANN, KNN SVM. The results showed were found be most influential prediction. validation Receiver Operating Characteristics (ROC) curve approach indicates area under AUC more than 0.82 for all models. best obtained using XgbTree with = 0.97 comparison MLA. Our findings demonstrate excellent ability ML models predict parameters. developed map valuable tool sustainable development, it can play effective role ensuring food security conducting land agriculture assessment.