Feature Importance in Decision Trees | Machine Learning Interpretability
In this video, we explain how to derive feature importance from decision trees. We explain how by understanding how a tree is constructed and evaluating the reduction in the impurity of the data obtained at each node, we can estimate a feature's contribution to the outputs of the model. We'll walk through an example using scikit-Learn and discuss how features used at multiple nodes or in higher nodes tend to have greater importance.
Want to learn more? Check out our course: https://www.trainindata.com/p/machine...