Comparative Analysis of Different Machine Learning Based Techniques for Crop Recommendation

Smart Agriculture is gradually becoming a blessing for mankind. It is very much efficient to mitigate food scarcity as well as minimize farmers’ efforts to produce sufficient food. Crop recommendation is one area which is helping the farmers to choose the best crop for a particular soil and climate. This study compares various machine learning techniques, including “Random Forest” (RDMFR), “Decision Tree” (DCTR), K Nearest Neighbor (KNNB), Radial Basis Function Support Vector Machine (RBFSVMN) and Radial Basis Function Neural Network (RBFNUNT) for recommending crops. Our objective is to check that potato and onion crops are suitable for a given soil and climate or not. We had checked the type of soil depending on “Nitrogen” (N), “Phosphorus” (P), “Potassium” (K), temperature, rainfall and moisture sensor data. An assessment of the performance of an machine learning (MLRN) classifier model is conducted using hybrid K cross-validation. Comparative examination of the entire system has been performed with the previous existing system.