Create Chatbot with Dialogflow Integration

Before getting started with the chatbot building, you need to perform training on the Dialogflow console. By adding a sufficient amount of training phrases, you can train the intent to recognize and react to more complex input from the end users.

For training intents on Dialogflow, you may refer to Dialogflow's documentation.

When you have completed setting up and training the intents, we can start creating the tree in WOZTELL in order to apply the intents to the chatbot.

Expected Outcome

The user's input matches with the "Weather Intent", as a result, an relevant response will be sent.

Sample Tree Structure

Create General Node: Greetings

First, create a general node to send a greeting message.

Create a response and ask for input from the end users.

Create Global Node: Entry Point

Create a global node as an entry point to this flow.

Create a new trigger that reacts to greeting keywords.

Switch on "Redirect" and link it back to the "Greetings" node. Now we have created an entry point.

Create Transformer Node

Create a transformer node following the first node. This is for detecting the intent from the user's input.

Create Trigger: Any Text

Create a new trigger with "Any Text", so that it can react to any text input by the end users.

Create Action: Dialogflow

Create a new action. Select "Dialogflow Detect Intent" from resource template. This action will help you with detecting intents from the user's input.