COMPARISON OF MACHINE LEARNING ALGORITHMS FOR LAND USE AND LAND COVER ANALYSIS USING GOOGLE EARTH ENGINE (CASE STUDY: WANGGU WATERSHED)

Human population growth and land use cover (LULC) change have always developed side by side. Considering selection of a good Machine Learning (ML) classifier algorithm is needed considering the high estimation LULC maps based on remote sensing. This study aims to produce classification Landsat-8 Sentinel-2 images comparing accuracy performance three ML algorithms, namely: Classification Regression Tree (CART), Random Forest (RF), Support Vector (SVM). Dataset comparison ratios were also explored find results with best accuracy. better than regarding Overall Accuracy (OA) Coefficient Kappa. The ratio training testing datasets level 70:30 both average OA being 92.09% 94.21%, respectively. RF outperforms CART SVM in types satellite imagery. mean CART, RF, classifiers was 92.03%, 94.74%, 83.54% Landsat-8, 93.14%, 96.15%, 93.34% Sentinel-2,