Model Options

In order to build the most basic model, you need to specify options on the Model tab before proceeding. Scoring method and other advanced options are available on the Expert tab.

The following basic modeling options are available:

Model name. Specifies the name assigned to the model that is created when the node is executed.

• Auto. Generates the model name automatically based on the target or ID field names or the name of the model type in cases where no target is specified (such as clustering models).
• Custom. Allows you to specify a custom name for the model created.

Use partitioned data. Splits the data into separate subsets or samples for training, testing, and validation based on the current partition field. Using one sample to create the model and a separate sample to test it may provide an indication of how well the model will generalize to larger datasets that are similar to the current data. If no partition field is specified in the stream, this option is ignored.

With Drillthrough. If shown, this option enables you to query the model to learn details about the cases included in the model.

Unique field. From the drop-down list, select a field that uniquely identifies each case. Typically, this is an ID field, such as CustomerID.