Create a workspace
- From the Watson Conversation tooling, open the conversation workspace, and click Create.
- Name the workspace
national-parks-example
, and then click Create. - Click the Dialog tab, then click the Create button.
The Watson tooling provides a “Conversation starts” virtual node and the first child node.
- In the Triggered by field, enter
conversation_start
. This condition is invoked on the start of a conversation. - In the Fulfill with a response field, add the following response:
What national park would you like to know about?
Note that the Conversation tooling automatically adds a new node with condition
anything_else
. This node is evaluated last and can be used as a failover when no previous node conditions match. - To create a failover message, enter a response to the node with condition
anything_else: I didn't understand that
. - Test the dialog that you’ve created so far by clicking the Chat icon.
When you open the chat, you see the first
conversation_start
node with a response ofWhat national park would you like to know about?
- Enter
How about Yosemite?
Notice that Watson doesn’t understand that intent and therefore fails over to the
anything_else
node for its response. - Add some variation to your responses. Start with the
conversation_start
node. - Click Create another response and enter an additional response variant:
I'm here to provide information about national parks. What park are you interested in?
- Click Create another response > Set to random to enable Watson to randomly choose one of the responses.
- Select the
anything_else
node, and modify the first response toI didn't understand that. Ask me about a national park…
. - Add another response variant:
Sorry I'm only trained on national parks. What park would you like to know about?
- Click Create another response > Set to sequential.
- To test the dialog, open the Chat pane.
- Click Clear to restart the conversation.
- Do this multiple times, and observe that the prompts are shown at random.
- Enter an utterance to test the
anything_else
node. For example, enterI'd like to know about Yosemite
.Notice that the responses are shown in sequence. First, you see the response
I didn't understand that. Ask me about a national park…
. - Enter another utterance. For example,
What about Arches
.Watson renders the second response as
Sorry I'm only trained on national parks
.