A research group mapped the forest cover of Indonesia based on combined Planet-NICFI and Sentinel-2 data. They benefited from the random forest approach. One of their objectives is to show how this method is simple and efficient for image classification as well as to achieve a high accuracy in classification by implementing high-quality image data. They evaluated their dataset to assess model performance and compared it with another forest cover dataset. 

Check out their research in more detail, and stay tuned for the upcoming user stories!