Page classifiers can be characters, barcodes, logotypes, keywords or anchor points that are used to help identify a document by identifying specific pages within the document.
Page classifiers are used to separate a stream of pages into individual documents by identifying the first—and possibly last—page of a document (using document separation).
After pages are separated into individual documents, page classifiers can be used to identify the document as a whole. They are also used to make sure that the data of interest is extracted from the correct pages of a document.
By definition, page classifiers are defined for structured documents (in the images that are added as document layouts in the document specification editor). However, they can also be specified for semi-structured information (in the main specification) if the document in question has multiple pages, and if you know which page they are found on (even if you do not know where exactly on the page they are found).
For more information, see: