Discovery Accelerator provides a Deduplication feature. The purpose of this feature is to minimize review, export, and production times by letting you identify and exclude similar items and duplicate items. Discovery Accelerator considers items to be similar when they have the same metadata properties, such as author display names, subjects, and number of attachments. Duplicate items have exactly the same content as well as the same metadata properties. The option to identify and exclude duplicate items is available only in cases that you have enabled for analytics. This is not true of similar items, however, which you can identify and exclude in all cases.
The Stack field on the Review pane lets you hide or show duplicate items and similar items in the review set
Note: If you sort the items before you select an option in the Stack field then, after you have selected the Stack option, the items may no longer be in the original sort order. For example, this is the case if you first choose to sort the items by the Author column and then select one of the Stack options.
To Enable Stack option in Review Pane the Case should be Analytic Enable. By default Analytic is Disabled.
Each DA Case has the option to enable Analytics to run within the scope of the Case.