An Artificial Intelligence-based Crop Recommendation System using Machine Learning

Agriculture is the backbone of the Indian economy and a source of employment for millions of people across the globe. The perennial problem faced by Indian farmers is that they do not select crops based on environmental conditions, resulting in significant productivity losses. This decision support system assists in resolving this issue. In our study, the AI system helps precision agriculture improve overall crop harvest quality and accuracy. This research feature selection, Industry 4.0, proposes one solution, such as a recommendation system, using AI and a family of machine learning algorithms. The data set used in this research work is downloaded from Kaggle, and labeled. It contains a total of 08 features with 07 independent variables, including N, P, K, Temperature, Humidity, pH, and rainfall. Then SMOTE data balancing technique is applied to achieve better results. Additionally, authors used optimization techniques to tune the performance further as smart factories. Cat Boosting (C-Boost) performed the best with an accuracy value of 99.5129, F-measure-0.9916, Precision-0.9918, and Kappa-0.8870. GNB, on the other hand, outperformed ROC-0.9569 and MCC-0.9569 in the classification, regression, and boosting family of machine learning algorithms.