Predicting Forest Cover Type using Decision Trees & Random Forest

Given only geographic and environmental variables, can we predict what type of tree grows in a 30 m x 30 m plot in a northern Colorado forest? To tackle this machine learning problem, I utilized exploratory data analysis, feature selection, and feature engineering to develop an effective Decision Tree model. Ultimately, ensemble methods such as Random Forest and “soft” voting proved to be the most accurate with this problem.

See the project on Github here

Leave a comment