Rice productivity is an important indicator of national food security, so an accurate analytical approach is needed to monitor and predict harvest yields spatially. This research aims to develop a rice productivity prediction model using remote sensing technology with Unmanned Aerial Vehicle (UAV) and Support Vector Regression (SVR) method. The study was conducted in Balumbang Jaya Village, West Bogor Regency, with a limited sample size of 30 rie field plots. Five vegetation indices were analyzed, including: Normalized Difference Vegetation Index (NDVI), Green Normalized Difference Vegetation Index (GNDVI), Normalized Difference Red Edge (NDRE), Optimized Soil-Adjusted Vegetation Index (OSAVI), and Leaf Chlorophyll Index (LCI). The research methodology integrated remote sensing techniques, multispectral image processing, and machine learning. The optimal SVR parameters were obtained through grid search with sigma=1 and cost=1. The Synthetic Minority Over-sampling Technique (SMOTE) was applied in initial data clasification stage to balance the productivity class distribution, although this study focused on regression. The results show that the SVR model with Radial Basis Function (RBF) kernel can explain 87.6% of rice productivity variability with a Root Mean Squared Error (RMSE) of 0.29 ton/ha. The findings confirm the effectiveness of a multidisciplinary approach in developing accurate and innovative rice productivity prediction models. This model has the potential to be used as a decision-making tool in agricultural land management that is more efficient and responsive to environmental variability