Soil Fertility Assessment and Crop Recommendation Using Machine Learning

Balancing the soil fertility is the crucial for sustainable agriculture since it helps maximize crop yields. The research adopts Machine Learning (ML) to construct a recommendation system for soil fertility and crop selection which examines soil nutrient changes preceding and following cultivation. Through forecasting soil fertility reduction the system provides crop waste recommendations for fertilization improvement. The system offers recommendations regarding suitable crops for the upcoming harvest cycle when using waste residues and when excluding waste residue use. The model receives training from authentic soil and crop data through Random Forest Regression and Classification methods. Through an application designed with Flask the system presents an interactive graphical user interface that enables farmers to quickly obtain recommendation outcomes by supplying soil parameters. Recovering the organic waste and facilitating precision agriculture through its operational framework, this technology increases the resource efficiency. Soil sustainability receives improvement through this method which simultaneously increases agricultural production and reduces chemical fertilizer dependence.