Use entities and slot filling

Task 01: Use entities and slot filling

Introduction

To enhance your agent’s ability to efficiently handle common inquiries from Contoso’s customers, you’ll import content directly from Contoso’s existing website. This task helps the agent leverage accurate, consistent, and already-approved customer service information, streamlining responses to frequent customer requests.

Description

In this task, you’ll use Microsoft Copilot Studio to import content from a website. Copilot Studio analyzes the webpage content and generates suggested conversational topics automatically, reducing manual content creation.

Microsoft Copilot Studio uses natural language understanding (NLU) to interpret what a user is saying to try and match a user’s utterance with an existing topic. NLU can also help the agent identify entities in a user’s input.

An entity represents key information you’re trying to extract from a sentence and then saves it for later use.

Example scenario

If the user enters “I want fifty red coffee machines” the agent can understand that:

“Fifty” is the number “50” and it’s also the number of products to purchase.
“Red” is a color and is the color of the products to purchase.
“Coffee machine” refers to the product that the person wants to purchase.

Success criteria

You’ve successfully imported topics from the provided website.
You’ve reviewed and confirmed the topics created from the imported content.

Key tasks

01: Create an entity

In this section, you’ll create an entity named Order Action.

On the command bar for the agent, select Settings.
On the Settings page, in the list of Settings, select Entities.
On the Entities page, select Add an entity and then select + New entity.
In the Create an entity dialog, select Closed list.
On the Unnamed entity pane, in the Name field, enter Order Action.
In the List items section, enter Update and then select Add.
Repeat Step 6 to add the following items:
- Check
- Cancel
Set Smart matching to on and then select Save.
Select Close.
Close the Settings page.

02: Update your topic to use the entity

On the command bar for the agent, select Topics.
Select the Check Order Status topic.
In the Question node, in the Identify section, select User’s entire response.
In the Choose information to identify dialog, search for and select the new Order Action entity.
Select Select options for user.
In the User options for Order Action dialog, select all three checkboxes and then close the dialog.
On the command bar for the agent, select Save.

By default, if the variable assigned to store the question’s response already contains a value, the question is skipped and not presented.

03: Test slot filling

To confirm the accuracy of information collection from customers, you’ll test the slot filling capabilities of your agent. This helps ensure that the agent correctly identifies and stores critical details from customer interactions for Contoso’s service requests.

In the Test your agent pane, select Start new test session to start a new conversation.
Submit the following prompt:
```
 Can I check on an order?
```
On the command bar for the agent, select Variables.
In the Variables pane, select Test and then select Topic.

You’ll see the process is working because the user has triggered this topic with the intent to check an order, and the entity has been slot filled into the variable from the follow-up question after the trigger phrase.

As a result, the question is not asked and is skipped. This is because you used entities and slot filling to retrieve the information from the utterance the user submitted.

You’ve successfully configured a custom entity for your Question node. By default, if the variable assigned to store the question’s response already contains a value, the question is skipped and not presented.