Skip to main content


         This documentation site is for previous versions. Visit our new documentation site for current releases.      
 

This content has been archived and is no longer being updated.

Links may not function; however, this content may be relevant to outdated versions of the product.

Verifying chatbot responses

Updated on July 22, 2021

To ensure that Pega Intelligent Virtual Assistant™ (IVA) responds correctly to user requests during a chat session, you can first simulate conversations with the chatbot and then verify the responses in the preview console for a conversational channel. As a result, the system becomes more responsive and improves its text analysis and artificial intelligence.

For example, you can ensure that the system correctly detects topics and entities relating to car insurance, when a user asks for a car insurance quote in their chat session.
Before you begin:
  1. In the header of Dev Studio, click the name of the application, and then click Channels and interfaces.
  2. In the Current channel interfaces section, click the icon that represents your existing Alexa, Unified Messaging, or Web Chatbot channel.
  3. In the channel, click the Behavior tab.
  4. At the top of the preview console, turn on the Show analysis switch.
    Result: When you enter sample text in the preview console, the analysis shows the topic detected and the system's confidence in the correctness of the detected topic, as a percentage.
  5. In the preview console in the right panel, type in a typical user phrase that starts a business case.
    For example: To request a car insurance quote, enter: I would like to request a car insurance quote for my Ford Mustang GT.
    Result: The preview console displays the detected topic below the text that you enter, and starts the associated business case in the IVA.
  6. Correct the topic (subject matter) or entity detected in user input:
    ChoicesActions
    If no topic was found
    1. Click Define topic below the entered text.
    2. In the Topic field, press the Down arrow key, and then select the correct topic, for example: insurance quote.
    3. In the field below the topic name, enter a sample of user input to associate with the topic.
      For example: Enter How can I get a car quote?
    4. Optional: To add additional sample phrases to associate with the topic, click Add example, and then in the field below, enter the text.
    5. Click Submit.
    If the wrong topic was found
    1. Click the topic name located below the entered text, for example: #new hire.
    2. In the Topic field, press the Down arrow key, and then select the correct topic.
    3. In the field below the topic name, enter a sample of user input to associate with the topic.
    4. Optional: To add additional sample phrases to associate with the topic, click Add example, and then in the field below, enter the text.
    5. Click Submit.
    If no entity was found
    1. In the text section below the detected sentiment, highlight and right-click the text that you want to map to the new entity.
    2. Click New entity.
      For example: To select a car make in the text, highlight the word Ford.
    3. In the Create new entity window, in the Entity name field, enter a name for the entity, and then click Submit.
    If the wrong entity was found
    1. In the text section below the detected sentiment, highlight and right-click the text for an existing entity.
    2. Click the name of another entity.
      To make sure the car make in the text maps to the CarMake entity, highlight and right-click Ford, and then click #CarMake.
  7. Click the Like icon next to the topic name located below the entered text.
    Clicking this icon is important in the IVA training process, and improves the correct detection of topics.
  8. Click Save.
For example: The following figure shows added entities in sample user input, in the preview console:
Two defined entities in text in the preview console
Car make and car model entities in user input in the preview console.
What to do next: Ensure that the IVA learns from the corrected topics and validated entities by building the text analytics model. For more information, see Applying simulation changes to a text analytics model.

Have a question? Get answers now.

Visit the Support Center to ask questions, engage in discussions, share ideas, and help others.

Did you find this content helpful?

Want to help us improve this content?

We'd prefer it if you saw us at our best.

Pega.com is not optimized for Internet Explorer. For the optimal experience, please use:

Close Deprecation Notice
Contact us