Applying Machine Learning Models to Identify Forest Cover

This paper presents an application of several classification techniques on forested lands data. More specifically, testing the efficiency of each classifier in its ability to identify a specific forest cover type. Four classifiers were used in the study, and testing was performed with both unscaled data and data scaled via two different methods. The study completes successfully with a stand-out algorithm that easily exceeded its peers in the given task; the Random Forest classifier. It concludes with speculation on how this algorithm, and the study itself, can be built-upon in the future.