Introduction to Watson Assistant

Lab 3 - Using digressions

Overview

This is a continuation of the lab exercise Using slots and server functions that introduces a new important feature of Watson Assistant.

Digressions - A digression occurs when a user is in the middle of a dialog flow that is designed to address one goal, and abruptly switches topics to initiate a dialog flow that is designed to address a different goal. With digression settings, you can allow the conversation to return to the dialog flow that was interrupted when the digression occurred. For example, the user might be ordering a new phone, but switches topics to ask about tablets. Your dialog can answer the question about tablets, and then bring the user back to where they left off in the process of ordering a phone. Allowing digressions to occur and return gives your users more control over the flow of the conversation at run time. They can change topics, follow a dialog flow about the unrelated topic to its end, and then return to where they were before.

in this lab you'll add a Digression to DinnerBot to allow the users to return restaurant hours wherever they are in the conversation flow.

Part 1 Adding digressions to DinnerBot

Step 1: Import the starting skill

If not open already, navigate to the Bluemix Console and launch your IBM Watson Assistant service.

To save time you'll import a skill as a starting point that has all the DinnerBot intents and entities already defined.

1. Download the saved skills file dinnerbot-digressions-begin.json to your local filesystem. (Click "Raw" from the link above to get the raw format to save locally)

2. From the Watson Assistant tooling click on the Skills tab and click on Create new

3. Click on Import skill, click Choose JSON file, select the file dinnerbot-digressions-begin.json and then click Import

4. The imported skill will open up in the tool. Click on the Intents tab and note that we've added an intent to request dinner hours for the restaurants. Take a look at the example for the #hours intent to see the examples that were used.

Step 2: Add the restaurant hours dialog nodes

1. Go to the Dialog, select the Book Reservation node and click Add node

2. Name the node Request Dinner Hours and select the #hours intent where it says if assistant recognizes

3. Click on Customize in the top right corner. Select Digressions, expand Digressions can come into this node, enable it, and then select Return after digression Click Apply

4. Select the node Request Dinner Hours and click Add child node. Name the node Restaurant set in context and enter the context variable $cuisine where it says If assistant recognizes

5. Click on Customize in the top right corner and enable Multiple responses. Click Apply

6. Add the following values to the area labelled Then respond with clicking on Add response after each entry

If assistant recognizes	Respond with
$cuisine:chinese	Our Chinese restaurant Hao Hao is open for dinner everyday from 6pm to 10pm
$cuisine:italian	Our Italian restaurant Osteria is open for dinner Mon-Sat from 6pm to 10pm
$cuisine:mexican	Our Mexican restaurant La Mesa Rosa is open for dinner Mon-Sat from 7pm to 10pm
$cuisine:french	Our French restaurant Le Cordon Bleu is open for dinner Mon-Sat from 7pm to 10pm
$cuisine:american	Our American restaurant Hemmingway's is open for dinner everyday from 7pm to 10pm

7. Select the new node Restaurant set in context, click on the 3 dots on the right, and then select Duplicate from the context menu

8. Change the name of the new node to Restaurant set in intent and enter the entity @cuisine where it says If assistant recognizes

9. In the section labelled Then respond with: change the values under If assistant recognizes to use the entity @cuisine instead of the context variable $cuisine as shown below (note: the responses don't need to be changed)

10. Duplicate the node you just modified and name it Restaurant not set. Enter anything_else where it says If assistant recognizes.

11. Click on Customize in the top right corner. Enable slots and select Prompt for everything. Click Apply

12. Add the following values to the area labelled Then check for for a single entry.

Check for	Save it as	If not present, ask
@cuisine	$hours_for	Which restaurant do you want the hours for?

13. Select the node Request Dinner Hours. Go to the section And finally and select Skip user input.

14. Select the node Book Reservation. Click on Customize in the top right corner. Select Digressions, expand Digressions can go away from this node, enable it and select Only digress from slots to nodes that allow returns Click Apply

Step 3: Test your chatbot

The Watson Assistant tooling offers a testing panel to test phrases to confirm the correct intents, entities, and dialog are matched and returned.

1. To test the bot, click on the Try it icon in the top-right corner of the tooling.

2. A side panel appears and shows the contents of the node that matches welcome. Enter a message that triggers the #hours intent. Start with what time does dinner start ?

3. Verify that you're prompted for a restaurant.

4. Enter Hao Hao or chinese and verify that the hours for the chinese restaurant are returned.

5. Click on the Clear link to start over with the test tool. Type When is dinner at the chinese restaurant when prompted.

6. Verify that the hours for Hao Hao are returned

7. Now you'll test a digression. Click on the Clear link to start over with the test tool. Type Book a table for 4 mexican next friday when prompted.

8. Verify that you are prompted for the time. Instead of the time enter what are the hours instead

9. Verify that the hours of the Mexican restaurant are returned and you're prompted for the time of your reservation again.

Summary

Pat yourself on the back. You've seen how digressions can be used to seamlessly handle certain types of user input at any point in the dialog flow.