Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.abbyy.com/llms.txt

Use this file to discover all available pages before exploring further.

The documents fall into two layout classes, each with its own set of Extraction Rules. The skill also needs to support standard Vantage training. Configure the processing flow accordingly.
1

Open the Activities tab

Go to the Activities tab.
2

Add a Classify By Text and Image activity

First, we need to determine the document class. Click the Classify By Text and Image activity on the Activities pane — it will be added to the workflow.When you add this activity, a new field is created in the skill structure to record the classification results. The field is hidden and not editable, but you can rename it in the Activity Properties pane on the Activities tab. Rename the field to “Layout”.
3

Add an IF activity to fork the flow

Documents should be routed to different activities based on classification results.
  1. Click the Classify By Text and Image activity in the workflow.
  2. In the pop-up window, select the IF item. It will be added to the workflow after the Classify By Text and Image activity.
Instead of using the IF activity, you can create a workflow item containing two Extraction Rules activities and select the “Layout” field as a source field so that its values serve as conditions for choosing which activity will be applied to each document. For more information about this option, see Several sets of Extraction Rules within a single activity.
4

Add Extraction Rules activities for each class

Add the activities for extracting data from documents of different classes.
  1. Select the Extraction Rules activity as the next item. Rename it to “Sick Note DE” — this activity will extract data from the German documents.
  2. Click the IF activity and add another Extraction Rules activity. Rename it to “Sick Note BE-NL”. This handles Belgian and Dutch documents (variants inside this class can be handled by a single activity).
The IF activity should now be connected to both Extraction Rules activities and to Finish.
5

Add a Fast Learning activity at the end

The Fast Learning activity enables further skill training during production via Online Learning.
  1. Click “Sick Note BE-NL” in the processing flow.
  2. In the pop-up window, select the Fast Learning activity. It is added after “Sick Note BE-NL”, but “Sick Note DE” is still connected to Finish.
  3. Hover over the arrow connecting “Sick Note DE” and Finish. The arrow turns orange.
  4. Drag and drop this arrow onto the Fast Learning activity.
  5. Hover over the arrow connecting the IF activity and Finish. Drag and drop this arrow onto the Fast Learning activity as well.
All documents are now forwarded to the Fast Learning activity.
Your final setup should look like this:
Document processing flow showing Classify By Text and Image, IF, Sick Note DE, Sick Note BE-NL, and Fast Learning activities
The exclamation marks indicate that the activities haven’t yet been configured. We’ll configure them in the next steps.

​
What’s next

Step 3. Configure and train the Classify activity

Upload training documents, assign classes, and train the classification activity.

Tutorial overview

Back to the tutorial introduction.