Crop recommendation system using machine learning

Abstract

Agriculture is the main field of employment in India. Farmers are faced with many problems when evaluating the yield of their crops. The production of crops plays an important role in our Indian economy. This proposed system helps the farmers choose suitable crops based on rainfall, humidity, type of soil, pH of soil, and temperature Accurate crop prediction results in increased crop cultivation. It will help farmers by reducing the losses they face and improving yield. Machine learning plays an important role in the area of crop cultivation. This work proposes a crop recommendation system using machine learning techniques such as k-nearest neighbor (KNN), artificial neural network (ANN), random forest (RF), and support vector machine (SVM). The models are simulated comprehensively on an Indian data set. The SVM predictive model had an accuracy of 97.85% and a training time of 218.691ms. The K-NN predictive model gave an accuracy of 97.95% a training time of 218.691ms, and the RF gave an accuracy of 99.22% a training time of 138.021ms. This model is beneficial to farmers because it allows them to know the type of crop before cultivating the agricultural field and thus encourages them to make suitable decisions.