After the model has been trained, the platform displays a chart with the 10 features that had the most significant impact on the model prediction of the target variable for the selected model. The chart is available in the Results tab.
The Feature Importance Matrix helps you evaluate the impact of features on the model. If a feature does not affect the model at all (normalized value of 0), you can exclude that variable from model building.
The FIM provides the following controls:
- Top 10/Bottom 10 — Shows the 10 most important or least important features.
- Select features — Lets you select or remove features of interest.
- Classes — Controls which classes are displayed on the bar chart (for classification task types only).