Document Separation and the Classification Set
There is no separation training document set in the Transformation Designer. The relationship between classification and separation is so close that the same information that is used to classify a document can also be used to perform separation.
For Standard Document Separation, the first page is recognized and a new document is created. For Trainable Document Separation, each individual page is classified. Based on the relationship between the individual page and the surrounding pages, an intelligent decision is made to separate the individual pages into the most logical documents.
If you enable document separation and confirm and the documents added to the Classification Set, the documents are used for separation.