Data Skeptic
Data Skeptic
Kyle Polich
[MINI] Calculating Feature Importance
13 minutes Posted Oct 21, 2016 at 9:24 am.
Download MP3
Show notes

For machine learning models created with the random forest algorithm, there is no obvious diagnostic to inform you which features are more important in the output of the model. Some straightforward but useful techniques exist revolving around removing a feature and measuring the decrease in accuracy or Gini values in the leaves. We broadly discuss these techniques in this episode.