Assign a class to documents

If you are using the Classification Set or the Extraction Set, the documents contained within those sets need to have an assigned class in order to aid in training. Any documents without an assigned class are not included in the online learning process until they are fully processed.

If you are using a benchmark set, an assigned class is required in order to compare your project classification settings against the assigned class when a benchmark is performed.

Important When you want to generate a classification or separation benchmark, first store the assigned class structure on disk using the Sort Documents on Disk by Class option from the Benchmark Sets shortcut menu.
Note If you assign a class to a document in a benchmark set and then convert that document set to a test set, the assigned class information is not lost. This information is stored in the XDocument (*.xdc), but it is not displayed for a test set. If you convert the test set back to a benchmark set, the assigned class is populated with its previous value.

You can assign a class to one or more documents by following these steps:

  1. Open the Documents window if it is not already open.
  2. Select the document set and the document subset with the documents that require class assignment.

    This can be either the Classification Set, the Extraction Set, or a benchmark set.

    The documents in the selected document subset are displayed in the selected view.

  3. If necessary, switch to the List View Documents Window - Flat View icon.

    The selected document set is displayed in a list.

    Note In the benchmark document set, you can also assign classes from the Hierarchy View.
  4. In the list of documents, select one or more documents that require class assignment.
    Tip To save time selecting individual documents, click Select All Select All icon or Ctrl + A to select all documents.
  5. On the Ribbon Documents tab, in the Document group, select Assign Class Assign Class icon. Alternatively select the option from the documents shortcut menu, or press Ctrl + H.

    An additional Assign Class pane is displayed next to the view.

  6. Select the desired class from the list by typing in the appropriate class name in the text box.

    The selected documents are assigned to the selected class. This means these documents are extracted using the locators and fields for the selected class.

    Tip You can remove an assigned class for a document by selecting no class.