Great work! In this lesson, you learned how to create decision trees and use them to make classifications. Here are some of the major takeaways:
- Good decision trees have pure leaves. A leaf is pure if all of the data points in that class have the same label.
- Decision trees are created using a greedy algorithm that prioritizes finding the feature that results in the largest information gain when splitting the data using that feature.
- Creating an optimal decision tree is difficult. The greedy algorithm doesn’t always find the globally optimal tree.
- Decision trees often suffer from overfitting. Making the tree small by pruning helps to generalize the tree so it is more accurate on data in the real world.
Sign up to start coding