Defining categories for automatic routing

When you set up a new Email channel as a Pega Intelligent Virtual Assistant, you can configure text analysis categories for the channel interface. Categories are a part of the selected taxonomy of a text analyzer rule and are used in natural language processing (NLP).

Categories are one of the methods used in the automatic routing capability to determine how to process the email content that was received from the user. You first define categories and later use them in routing conditions to help the Pega Intelligent Virtual Assistant make an informed guess about what to do with the received email based on its content. In a routing condition, you can specify to assign the email content to an operator, add it to a work queue, or create a top-level case.

You can also update the NLP model used for the configured Email channel so that subsequent text analysis results will have a better confidence score, and the entity extraction, category detection, and sentiment analysis will be more accurate.

You can add, delete, or modify each category definition in the list.

Note: You must purchase a separate license before using Pega Intelligent Virtual Assistant for Email in your application.
  1. Open an Email channel interface.
  2. Click the Text analysis tab.
  3. Define a category.
    1. Click + Add category.
    2. Fill out the following fields:
      • Category - Required. The text analyzer category for the Email channel to use as the response based on the Approximate match, Must match, and Never match criteria of the received email content.
      • Approximate match - Optional. The comma-separated list of words or short phrases to use as an approximate match for the received email content that triggers the action assigned for the category.
      • Must match - Optional. The comma-separated list of words or short phrases to use as an exact match for the received email content that triggers the action assigned for the category.
      • Never match - Optional. The comma-separated list of words or short phrases to match in the received email content that will never trigger an action for the category.
  4. Optional: Repeat step 3 to define additional categories.
  5. Click Save.