Crop classification and prediction based on soil nutrition using machine learning methods

In India, farmers generally grow a traditional crop or the crop that is in demand resulting in poor yield. In the former case, the nutritional value of the soil gets deteriorated due to the non-rotation of crops, while in the latter the soil nutrient is not analyzed for the suitability of the new crop. As a result, people may suffer from stress and depression due to low income. Taking these into account, for classifying and predicting the suitable crop based on the soil nutrition levels a model is proposed using machine learning models such as Decision Tree, Support Vector Machine, Naive Bayes, K-Nearest Neighbours, Extreme Gradient Boosting and Random Forest. The dataset for this research is collected from the Kaggle website consisting of 6 different crop types with 11 nutrients. The models are trained and tested with 80% and 20% of the dataset respectively. The results prove Extreme Gradient Boosting followed by Naive Bayes to perform better with an AUC score of 0.994 and 0.993 respectively when compared to other models. © 2023, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer Applications and Management.