Data overview

The data overview feature is available on the Data tab of all applications, on the Modeling tab and the Score tab in IBM® SPSS® Modeler Advantage, and on the Deploy tab of applications that include the Score Now feature.

1. To run an overview of a data source, click the Data Overview icon available throughout the application.

   

2. The Data Overview dialog will appear. If desired, select an overlay field from the drop-down in the Overview Options section.

   Then after running the overview, tabs will be available to display results for the primary selected field only, or to overlay the primary selected field with the field specified in this drop-down. For example, in the results you may want to view information about the Age of customers, and then overlay it with another field such as Gender.

3. If desired, select Use partitioned data in the Overview Options section. This option is available if, under Optional Settings on the Modeling tab, you selected the option Automatically partition data to enable model evaluation on build data source for evaluation and testing. This option splits the data into separate subsets or samples for training and testing the model. By building the model on one subset and testing it on another, you can get an idea of how it will generalize to other data sets.
4. Select the data fields to include in the overview and click Run Overview. All fields available in the data source are listed. By default, all model input fields and the target are selected.

   The data overview will run and the results will appear. You can sort the information and choose which columns to display.

5. Click any field to see its details. A new results tab will open for each field you select, allowing you to view charts and tables and select overlay fields, if available.