The Dry Chaco Forest is a key forest stratum in Paraguay, representing 56.3% of the country’s forest cover and hosting 77% of the national changes in native forest land use. This stratum is crucial for monitoring forest cover, and its conservation is vital due to the high rate of land use changes observed. This work presents a methodology for land use classification using machine learning techniques, utilizing Landsat-8 satellite images to analyze changes that occurred between 2018 and 2022. The first phase involves the preprocessing of satellite images, enhancing their quality with filtering and pansharpening techniques. In the second phase, machine learning algorithms such as Random Forest, K-Nearest Neighbor (KNN), and Support Vector Machine (SVM) are applied for classification. The results indicate that Random Forest offers balanced performance, while SVM is particularly effective in detecting specific changes. KNN, although effective, showed a slight disadvantage in identifying areas of change. The proposed methodology facilitates the acquisition of updated data and can contribute to better management of the Dry Chaco Forest.