Power Virtual Agents #9: Entities for Conversation in chatbot


Natural language understanding is the ability for Power Virtual Agents to understand what the user is telling it. The bot is able to route the user to the topic related to, even if that exact phrase isn’t listed as a trigger phrase.

An entity represents a real-world subject, such as a phone number, zip code, city, or even a person’s name. Your bot can smartly recognize the relevant information from user input and save it for later use

There are two types of entities:
Pre-built entities represent the most commonly used information, such as age, colors, numbers, and names. Power Virtual Agents bots recognize these automatically.
Custom entities are entities you make. While the pre-built entities cover commonly used information types, sometimes you’ll need to teach the bot’s language understanding model some domain-specific knowledge. For instance, you may need to create a custom entity for your product types

Create Custom entity

  1. Go to https://powerva.microsoft.com/
  2. Select the bot and Environment from the top right corner.
  3. Click on left pan on Entities and Click on New entity.

4. Give the name of the entity and List Item to the entity.

5. Each item needs to add the Synonyms.

6. Click on Synonyms, Add the Synonyms and Click on Save.

7. Make sure the smart matching toggle is set On. Click on save the entity and Close the the dialog.

Topic Creation with the New Entity

8. Click on Topics on left pane and Click on the New topic from top bar.

9. Give the Name for Topic and Give the Trigger Phrases.

10. Add the Question node to it.

11. Add the Question node to it to add the custom entity to ask the Color.

12. Now add this entity items in the condition nodes and click Save.

Test The Topic

Enter MackBook in the massage box and click send.

Select the option from Question Node.

Enter the color from the item from custom entity. as color is Red.

Hope you learn something, Thanks for Reading 📖!

Happy Power365ing!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.