Leveraging the Power of MRMR in Machine Learning Models for Multi Class Classification of Rice to Promote Sustainable and Efficient Smart Farming

Rice variety classification is of significant importance in the agricultural domain since it allows for effective crop management, quality evaluation, and yield optimization. This research paper presents an intelligent system for automatic rice variety identification into multiple classes using machine learning techniques. The Maximum Relevance Minimum Redundancy (MRMR) attribute selection technique is used in the framework to discover the most important attributes from a large dataset, ensuring accurate and reliable classification. Various machine learning based classification techniques, including Decision Trees, K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Ensemble methods, Neural Networks (NN), and Naive Bayes, are explored in their different variants. Series of experiments were conducted on a real-time dataset featuring multiple rice varieties to evaluate the performance of each classifier based on metrics such as accuracy, precision, recall, and F1 score. The study explores the effectiveness of the proposed framework, revealing that Ensemble machine learning, SVM and Neural Networks emerge as the optimal classifiers, achieving an impressive accuracy rate of 99.8% in the multi-class classification of rice varieties. The proposed framework empowers farmers and researchers to make informed decisions in crop management, resource allocation, and ensuring food security in agricultural practices.