Now that you have a relevant dataset, you must choose the target feature that you want to predict.
Select the target feature to predict
To choose the target feature that you want to predict:
- Review the list of available features in the Select Target view.
Note: Feature summaries show the values for each column/field in the dataset. Distil automatically infers the type of data stored for each feature. To change the inferred value, use the type drop-down on the upper right corner of the summary.
- Use the Search field to filter on event.
-
On the Event_Type feature summary, click Select Target.
Distil automatically sends you to the next step, adding the feature to the model definition and listing a sample of actual event type values.
Note: You can change the target at any time by returning to the Select Target view. Click New Model: Select Target in the navigation bar to go back.
The target you select determines the type of results the model will generate.
Feature type | Model type | Prediction | Results |
---|---|---|---|
Continuous | Regression | Numerical value | Correct and incorrect predictions based on a configurable distance from the actual target feature values |
Categorical | Classification | Category | Correct and incorrect predictions based the actual target feature category |
Timeseries | Forecasting | Numerical values over time | Forecasted values for the last 10% of timeseries data |
Event_Type is a categorical feature that will produce classification results.
Next steps
Select the features that the model should use to predict the target feature and begin the model generation.