COMPARATIVE ASSESSMENT OF MACHINE LEARNING MODELS IN RICE CROP STAGE CLASSIFICATION

Abstract

The effective classification of rice crops plays a crucial role in optimizing agricultural management and enhancing yield forecasts. In this paper, we explored the efficacy of various machine learning (ML) techniques in advancing the classification of rice crops. Four machine learning classification algorithms, namely k-nearest Neighbor (KNN), Support Vector Machine (SVM), Random Forests (RF), Decision Trees (DT), and XG Boost, are assessed using a dataset comprising rice crop images and environmental parameters. The studys findings reveal that XG Boost significantly outperforms other models, achieving an impressive accuracy of 96.78%, along with high precision and F1-Score. The Support Vector Machine also demonstrates strong performance with an accuracy of 93.83%. These findings emphasize the potential of ML in advancing agricultural practices and decision-making, highlighting the role of precision agriculture in food security.