Defining the Parameters Used to Train Predicted Attributes

Parameters are the training features used to train the predicted attribute.

Specify a minimum of one and a maximum of 10 parameters. The predicted attribute is trained based on the values of the specified parameters.

To specify the parameters for a predicted attribute:

1. Navigate to the Parameters tab of the Create Predicted Attribute or Edit Predicted Attribute dialog.

   By default, the account segments are added as parameters. Additional parameters can be included based on the Target selected for the predicted attribute. Delete the parameters that are not required.

2. Click Add to create a new parameter. A new row is added to the table.
3. In Name, select the attribute that must be used as a parameter. The Type is populated based on the attribute selected.
4. Click Ok to save the predicted attribute definition.

Note:

If the list of parameters is modified after the predicted attribute is trained, you must retrain the model so that it reflects the changes made.