Smart Mapping helps you instantly understand the key drivers in your data.

It is a supervised machine learning routine that enables you to select a target data feature and a set of additional features to be compared, then outputs those additional features sorted by their relative importance to the target. This helps you easily see the level of impact other features have on your target feature.

Smart Mapping also creates several AI-suggested visualizations for you to further interpret the results. This is especially useful in minimizing time to insight when working with a large number of features in your data. 

Using Smart Mapping

1. Click the Smart Mapping icon ( ) in the top toolbar to open the Smart Mapping panel.

2. Click and drag a feature from the Features panel to the Add Target field. 

   Tip: Continuous and discrete numerical and categorical features all work well as the Smart Mapping target.

3. Add other features by dragging and dropping them from the Features panel into the Add Features section. Or, to add all Features, click the Input All button. Choosing Input All will not input an already-selected target feature.

4. (Optional) Remove unwanted features by hovering over the feature and clicking the remove button ( ). 

   Tip: Remove features used to calculate the target and sparse features (with many missing values). Click the red pound sign to remove sparse features ( ).

5. Click Run at the bottom of the panel.

Reviewing Smart Mapping Outputs

Once you've run the Smart Mapping routine, you'll see a number of outputs:

Insights

Upon Smart Mapping's completion, a Notification will appear to let you know if Insights have been identified. Simply click View Insights in the notification to open the Insights Panel. Alternatively, you can click Show Me Later to view Insights at a later time.

Feature Impact

The Smart Mapping panel will update to show the other features you selected and the level of impact on your target feature. This level of impact is shown as a bar chart under each feature.

Note: Linear and non-linear relationships are explored and leveraged in determining the rank of the features impact on the target feature.

Additionally, you can toggle between Sort By Ranking and by Correlated Groups by clicking the Menu icon (  ) at the top right of the Smart Mapping panel.

• Sort By Ranking: Lists features from highest to lowest impact on the Smart Mapping output from top to bottom.
• Correlated Groups: If Smart Mapping identifies correlations between your features, these groups will be displayed with the impact of those features on the Smart Mapping listed from highest to lowest within the groups. Any un-correlated features will be listed below.

Suggested Mappings

At the bottom of the Smart Mapping panel, you'll see additional AI-suggested visualizations to help you interpret your data. The suggested mappings consider the relative importance of the features and any correlations existing between them. Simply click on the left or right arrows to switch between and click Apply to update the mapping.

Default Mapping

Upon running Smart Mapping, Mapping View will automatically populate with a recommended visualization to help you interpret your data:

• If the target is a numerical feature, it will be applied to the color dimension and split into quartiles.
  

  Tip: You can change the color bin options by clicking the Color Dimension Settings icon ( ) in the Mapping panel.

• If a categorical feature is within the first eight ranked features, it will be applied to Playback.
• The top binary categorical feature will also be mapped to Shape.

You can leverage the options in the Mapping Panel to adjust these items, any of the plots, or additional attributes.