Dialogflow CX tutorial for rookies Half-2

What’s webhook?

A webhook is a consumer outlined HTTP callback that’s mechanically invoked each time sure standards is fulfilled. A webhook might be created in any server-side programming language like Python, PHP or Node.js. You possibly can learn extra about webhook right here.

In Dialogflow, a webhook can be utilized to fetch information out of your server each time a sure intent having webhook enabled (don’t fear, we are going to see that) is invoked. The knowledge from the intent is handed to the webhook service to obtain the end result.

We’re going to use Ngrok, (an online tunneling instrument) that can be utilized to name webhook out of your native server. You possibly can obtain Ngrok from ngrok.io.

Let’s begin constructing our personal Dialogflow agent with a webhook utilizing Python (Flask) and deploy it utilizing Ngrok.

Set up Python

Obtain and set up python from right here as per your supported OS.

Set up Flask utilizing PIP

Flask is a light-weight internet framework that can be utilized to create webhook companies which might talk with exterior functions (in our case Dialogflow’s agent). To make use of flask in our app, first we have to set up it. It may be simply put in utilizing “pip”. If “pip” is just not put in then you have to to put in it first.

pip set up Flask

For extra data, you may go to: https://pypi.org/project/Flask

Creating webhook utilizing Python

Let’s create a fundamental flask app.

# import flask dependencies
from flask import Flask


# initialize the flask app
app = Flask(__name__)


# default route
@app.route('/')
def index():
   return 'Hey World!'


# create a route for webhook
@app.route('/webhook')
def webhook():
   return 'Hey World!'


# run the app
if __name__ == '__main__':
   app.run()

Now, run the app utilizing the next command.

python app.py or FLASK_APP=whats up.py flask run

Our fundamental app is working now, however as it’s on the native system it can’t be accessed from the surface world. To combine it as a webhook for Dialogflow, we have to make it dwell/on-line. For that, we use ngrok.

Run the webhook utilizing Ngrok

Ngrok is an online tunnelling instrument that gives a technique to take a look at APIs and webhooks from an area server. There are two variations out there. We shall be utilizing the free model with registration on ngrok.io.

To run ngrok, use the next command: ngrok http <port_number>

e.g. ngrok http 5000 (for flask app)

Easy methods to setup webhook in Dialogflow

Dialogflow supplies an choice on the left sidebar often called Webhook. Simply enter your webhook URL and webhook identify generated by ngrok and you’re completed.

You too can customise Webhook timeout.

You’ll want to add the URL with /webhook like…

https://41f5-2401-4900-1f3e-8fff-00-10d-c052.ngrok.io/webhook

and never simply

https://41f5-2401-4900-1f3e-8fff-00-10d-c052.ngrok.io/

as we’re going to deal with the request on /webhook route not the index route.

For those who present a URL with out webhook then chances are you’ll get an error like “Webhook name failed. Error: 405 Technique Not Allowed.”.

Enabling webhook for pages

Now, we have to allow the webhook for pages that want to speak with our server information. To do this, parameters from the web page for which you wish to allow the webhook, scroll all the way down to the top and allow the “webhook” from the webhook setting dropdown choose the webhook and fill within the Tag field which you wish to give the tag.

Each time this web page is invoked, it is going to ship a request to your webhook and reply as per the response set from the webhook.

Constructing success responses from webhook

Webhook responses ought to be in a correct JSON format in order that Dialogflow can perceive what to show to the consumer.

Under code will put together a easy JSON response with success textual content for Dialogflow. The consumer will obtain “This can be a response from webhook.” as response for this intent.

# import flask dependencies
from flask import Flask, request, make_response, jsonify


# initialize the flask app
app = Flask(__name__)
# default route
@app.route('/')
def index():
   return 'Hey World!'


# create a route for webhook
@app.route('/webhook', strategies=['GET', 'POST'])
def webhook():
   return make_response(jsonify(cx_response()))


def cx_response():
   information = request.get_json(silent=True, power=True)
   tag = information["fulfillmentInfo"]["tag"]


   if tag == "Welcome":
       reply = {
           "fulfillmentResponse": {
               "messages": [
                   {
                       "text": {
                           "text": [
                               'This is a response from webhook.'
                           ]
                       }
                   }
               ]
           }
       }
       return reply




# run the app
if __name__ == '__main__':
   app.run()

You possibly can see that we now have fetched “tag” from the request utilizing

tag = information["fulfillmentInfo"]["tag"]

In our instance, we now have used a tag, you should use it to your objective.

Checking response from webhook

Utilizing the console from “Check Agent” on the appropriate facet of the window, we will invoke the intent and examine the response. For our instance, the response shall be as beneath:

Utilizing this tutorial you may create a fundamental webhook. 

Hope this tutorial shall be useful to you in integrating webhook in your dialogflow app. If in case you have any questions then put up it within the remark. We’ll reply to it quickly.

What’s Occasion Handler?

In Dialogflow CX, an occasion handler is a kind of dialog node that lets you set off particular actions or responses based mostly on an occasion that happens throughout a dialog. An occasion might be regarded as a sign or set off that signifies one thing has occurred or a situation has been met.

If you create an occasion handler in Dialogflow CX, you specify the identify of the occasion that it ought to hear for. When that occasion is triggered, the occasion handler is activated and may carry out a lot of actions, comparable to setting context, sending a message, or invoking a webhook.

Occasion handlers are generally used to deal with conditions the place the consumer supplies particular data or triggers a selected motion, comparable to deciding on a selected merchandise from a listing, confirming an order, or asking for assist. By utilizing occasion handlers, you may create extra complicated conversational flows that reply to particular consumer actions or situations.

Easy methods to set an Occasion handler?

1. Open your Dialogflow CX agent within the Dialogflow CX console.
2. Click on on the “Circulation” tab to open the visible circulation editor.
3. Within the circulation editor, choose the dialog node the place you wish to add the occasion handler.
4. Within the node settings panel on the appropriate, scroll all the way down to the “Occasions” part and click on the “Add Occasion” butt- Within the “Add Occasion” dialog, choose the occasion that you simply wish to set off the occasion handler.
5. Optionally, you may add any further parameters or situations for the occasion handler.
6. Click on “Add” to create the occasion handler.
7. Now you can configure the actions that you really want the occasion handler to carry out, comparable to sending a message or invoking a webhook. You are able to do this by including actions to the node’s response part or by including a webhook to the node’s success part.
8. Save your modifications to the dialog node.

Constructed-in occasions:

The next occasions are built-in and are invoked by Dialogflow. Some occasions are restricted to sure ranges.

You possibly can confer with extra built-in occasions at this hyperlink https://cloud.google.com/dialogflow/cx/docs/concept/handler#event-built-in

What’s an entity?

An entity represents a selected object or idea that you really want your chatbot or digital assistant to have the ability to acknowledge and reply to. Entities are used to extract necessary data from consumer enter and to assist your bot perceive the which means of what the consumer is saying.

An entity generally is a easy worth, comparable to a quantity or a date, or a extra complicated worth, comparable to a product identify or a location. When a consumer inputs a message, Dialogflow CX analyzes the textual content and extracts any related entities which are talked about. For instance, if a consumer asks “What’s the climate like in New York?”, Dialogflow CX can extract the entity “New York” as a location.

Entities in Dialogflow CX might be both system-defined or user-defined. System-defined entities are built-in entities which are supplied by Dialogflow CX, comparable to “@sys.date-time” or “@sys.quantity”. Consumer-defined entities are entities that you simply outline your self, based mostly on the precise wants of your chatbot or digital assistant. You possibly can create a user-defined entity by defining a listing of doable values or by utilizing common expressions to match patterns in consumer enter.

After you have outlined an entity in Dialogflow CX, you should use it in intents to match consumer enter and extract the related data. You too can use entities to set parameters in success or to generate dynamic responses based mostly on consumer enter. By utilizing entities successfully, you may create a extra pure and conversational expertise to your customers.

Customized entity:

You possibly can create customized entities to acknowledge and extract data from consumer inputs. Customized entities are used to characterize particular ideas or objects which are related to your conversational AI software.

To create a customized entity in Dialogflow CX, comply with these steps:

1. Within the Dialogflow CX Console, go to the Entities web page.
2. Click on on the Create Entity button.
3. Within the Create Entity dialog field, enter a reputation to your entity.
4. Select a kind to your entity. There are a number of sorts out there, together with:
   • Customized: A customized entity that you simply outline.  
   • System: A predefined system entity that acknowledges widespread ideas comparable to dates, instances, and numbers.    
   • RegEx: A daily expression entity that matches patterns in consumer enter.

5. Choose the language that your entity shall be utilized in.
6. For those who select Customized as the kind to your entity, outline the values that your entity can acknowledge. You possibly can enter particular person values, or you may import a listing of values from a CSV or JSON file.
7. Click on on the Create button to create your customized entity.

After you have created your customized entity, you should use it in intents and fulfillments to extract data from consumer enter and use it to supply related responses.

Be at liberty to remark your doubts/questions. We’d be glad that can assist you. In case you are searching for Chatbot Growth or Pure Language Processing companies then do contact us or ship your requirement at letstalk@pragnakalp.com. We’d be pleased to supply our professional companies.