Fixing classification issues

After you upload batches of documents, IBM Automation Document Processing automatically recognizes the document type and you can fix issues that might have occurred during processing.

About this task

If the status of a batch after processing is Document issues, it means that the batch encountered problems when selecting the document type for one or more documents within the batch. Some problems that contribute to classification issues can include the following examples:

  • Some pages are not ordered correctly within a document.
  • A document contains pages that should be split into separate documents. For example, a document contains pages that correspond to different document types: the first page is an invoice, the second page is a purchase order, and so on.
  • Pages in different documents should be merged into one document.

To proceed with the document processing, you must first fix the classification issues.

Note: At the beginning of the document list, you see a warning and the number of issues in the batch. Issues are dismissed when you do one of the following actions:
  1. When you modify the content of the document.
  2. When you modify the document type.
  3. When you click Dismiss.

Procedure

  1. In your batch document processing application, select the tile Document type and page order to see all batches that have issues.
  2. Select the first batch.
    You see the list of documents in your batch. When you are working with a batch, it is automatically locked to prevent conflicts.
  3. For each document that has a warning, click Change document type edit icon and update the document type. You can choose from a list of pre-trained document types, and the system can recommend to you the most likely types for a specific document.

    You must select a document type among the list for the batch to be processed, even if some documents do not seem to match any of the available types. If a document does not correspond to any type in the list, and you do not need to process this document, you can delete it.

  4. Review the pages within your documents. Use any of the following options to fix the layout of your document:
    • Select one or more pages to rotate clockwise rotate clockwise icon, counterclockwise rotate counterclockwise icon, or 180 degrees rotate 180 degrees icon.
    • Select one or more pages and copy copy icon, cut cut icon, and paste paste icon them within the same document, or in a new document.
    • From the thumbnails pane, drag and drop pages to reorder them.
    You can use the following options to improve your own readability of documents. These settings are not saved to the documents:
    • Select one page at a time, and click visual effect icon to adjust brightness, contrast, and luminance. You can click Reset to revert to default settings.
    • Select a view:
      • Page view page view icon (full page)
      • Thumbnails view thumbnails view icon
      • Page and thumbnails view page and thumbnails view icon
    • Adjust the page size to fit to window fit to window icon, window width fit to width icon, or window height fit to height icon.
    • Zoom on the page:
      • Click Show magnifier magnifier icon to use a movable magnifier to enlarge part of the viewer. Click the icon again to disable the magnifier. Use the following keyboard shortcuts to adjust the magnifier:
        Magnifier action Key
        Increase size + or Shift+=
        Decrease size - or Shift+-
        Increase width Shift+Right Arrow
        Decrease width Shift+Left Arrow
        Increase height Shift+Up Arrow
        Decrease height Shift+Down Arrow
        Increase factor * or Shift+0
        Decrease factor / or Shift+9
      • Click Zoom area zoom area icon to enlarge a specific portion of the page.
      • Use the zoom in and zoom out buttons to zoom in and out.
        • When the Page and thumbnails view is visible, left-click the zoom buttons to zoom in or out of the main page, and right-click the buttons to zoom in or out of the thumbnail.
        • When Page view or Thumbnails view are visible, zoom options apply to the element that you can see on screen.
  5. Optional: If you must split a multi-page document into several documents, click Add to create a document in this batch and paste the pages that you cut from the original document.
    Note: This action is available only for PDF and TIFF documents.
  6. If you created a new document, you must set its document type from the list. By default, the document type is inherited from the document that is currently open.
  7. To save your changes, for example if you need to interrupt a session but do not want to submit the documents yet, click Save changes. If you want to cancel all changes instead, click Cancel.

    The documents are reprocessed by the system when you save your modifications, regardless of whether they have a document type or if the type is undefined.

  8. When you finish your changes and are ready to submit, click Submit.

Results

The batch is reprocessed with the correct classification, and you can then go on to fix classification issues for another batch. If after being reprocessed, the status of your batch is Data issues, you must go to Data extraction issues to verify extracted data. For more information, see Fixing data extraction issues.
Note: If you have an empty document in a batch, an error is shown during processing. You must either fill this empty document by pasting pages in it, or delete this document.