Leveraging Machine Learning and Spatial Analysis to Analyze Dynamic Land-Use Changes in Indramayu Regency to Enhance Food Security

Indramayu Regency, a national food barn, has determined sustainable agricultural land to bolster national food security by prohibiting land-use conversion. Studies on land use and land cover changes are essential to manage land cover changes, particularly in sustainable food agricultural areas, and to support food security. This study aims to classifiy land cover in Indramayu Regency using Random Forest and Support Vector Machine, and spatial analysis to identify land cover change trends. The study was conducted on Sentinel-2A satellite imagery from 2019 to 2023 using the Google Earth Engine. The results indicate that Random Forest achieved higher accuracy than SVM, with the highest kappa value of 0.9 and overall accuracy (OA) of 93%. Spatial analysis revealed that the most significant land cover change occurred in paddy fields converted to bare land, covering an area of 4301.33 ha, or 860 ha per year. In addition, paddy fields were converted into residential areas, encompassing 1810.35 ha, or 362 ha per year. The use of machine learning algorithms, particularly Random Forest, proved to be effective in analyzing land cover changes in the Indramayu Regency and can contribute to policymaking to enhance food security.