Population growth and economic and industrial development coupled have significantly accelerated the rate of Land Use and Land Cover (LULC) changes, particularly in developing countries, so finding optimum ways to observe these change has become a pressing issue. Quantification evaluation of these changes is crucial to comprehend and oversee land management conversion, therefore, it is necessary to evaluate the accuracy of various algorithms for LULC classification to determine the most effective classifier for Earth observation applications. The performance of Maximum Likelihood (ML), Support Vector Machines (SVM), Random Forest (RF), and K-Nearest Neighbors (KNN) was examined in this study, based on Sentinel 2A satellite images. The accuracy of those classifiers was evaluated using the Kappa Coefficient and normalized difference index-based verification. The findings indicate that all classifiers exhibit high accuracy levels with variations. The RF algorithm had the highest Kappa coefficient of 0.90, while the KNN algorithm the lowest of 0.76. The accuracy values for RF, SVM, ML, and KNN were 93.1%, 91.2%, 86.2%, and 82.5%, respectively. Results from this study using index-based LULC show that the RF classifier outperforms the others. The results of this study can be used in monitoring LULC change tasks.
