Creating a Scoring Catalog

To create a new scoring catalog in Predict, first click the Scoring Catalogs option in Predict, then click the Create New Catalog button.

This will open the New Scoring Catalog dialog.  Provide a name and an optional description for your new catalog.  You can also specify into which Dataset Library the scoring catalog will be placed.  By default, scoring catalogs become datasets in the "Scoring Catalogs" dataset library.

Click OK when ready to continue.  The Scoring Catalog Settings dialog will open.

A unique primary key(s) must be defined in the scoring catalog. To do this, select a template dataset and the unique primary key(s). For example, CustomerID or ProspectID are common unique key names. Sometimes there is a combination of variables that identify the uniqueness of a file, such as CustomerID combined with OfferType.

Use the Template Dataset dropdown to locate a file with the unique primary key(s). Often, this will be the dataset on which scoring jobs will run, but it does not necessarily have to be.  Then, select the unique primary key(s) using the Primary Keys drop down. Most often, there is only one variable, but you may select as many primary keys as necessary to ensure uniqueness.  Note that the primary keys of a scoring catalog cannot be changed once they are set.  If you have made a mistake or need to create a new catalog with different keys, you can easily create or delete any number of catalogs. 

Click Save when done.  You will then be presented with an opportunity to set permissions for this catalog.  You can find more information on the permissions dialog here: