Analysis and Prediction of Soil Nutrients Ph,N,P,K For Crop Using Machine Learning Classifier: A Review

Abstract

Agriculture is representing as an essential element in developing areas like India and other countries. The area of agriculture, computer engineering can be used for farmer decision-making for better yield and crop, also for soil quality purposes. Machine learning classifiers have a significant role in the determination of making distinct problems like soil nutrients and plant disease-related. In this paper, we have discussed the role of machine learning in the agriculture sector and also discuss distinct machine learning classifiers and their associated work in soil nutrients in the context of the agriculture sector. This paper provides a survey of various machine learning classifiers that include J48, Naive Bayes, k-NN, Random Forest, SVM, and JRip for crop prediction in the respective region according to the available soil. Most of the paper has proposed their work by considering sulfur, iron, EC, boron, OC, zinc, copper, pH, nitrogen, phosphorus, magnesium, potassium, etc., but the focus will be mostly on pH, N, P, and K. Naive Bayes classifier gives good result as related to other classifiers of machine learning used for large data set. © 2021, Springer Nature Switzerland AG.