Using Third party Tools to create Chatbot in PeopleSoft Part 3

This is final post to demo the working chatbot in PeopleSoft.

Kommunicate supplies you with HTML code to add their chat badge on your PeopleSoft pages, I added it to Fluid Homepage. You can also pass the system variable like User ID to Kommunicate chat doing this you can authenticate user in the simple manner.

In below image user clicks on the chat badge, as soon as user clicks on the badge our first 'welcome' event is fired from Dialogflow, Dialogflow sends the web service request to Django App, Django App sends web service request to PeopleSoft, PeopleSoft returns the response with User Name which is why you are seeing 'Welcome User! What can I help you with today?'

 

User types the question 'I need help', our second event is executed we again send a response in JSON format, Kommunicate will display your message accordingly with clickable buttons. 

User clicks on 'My Approvals' button, PeopleSoft send below response

User clicks 'Approver xxxxxxxxx' voucher

We approve the voucher or call any workflow in place and display appropriate message.


You can integrate Dialogflow with PeopleSoft by removing Django but it will limit your NLP processor to only work with PeopleSoft, Django will remove this limitation. Django will enable your chatbot to work with any application, not just PeopleSoft.

Using Third party Tools to create Chatbot in PeopleSoft Part 2

Now that Dialogflow setup is complete which will act as your NLP, it is time to write your Django application which will act as a webook from dialogflow.

It is super simple to create your first Hello World! app, visit and try creating a sample app https://www.djangoproject.com/, once you get familiar with Djangoyou are ready to go, below is an example of view I created in Django application which will provide webservice.

This screenshot is from my views.py file in Django Application. Below code is basically a providing a webhook/fulfillment response to  Dialogflow NLP (Refer to part 1 step 7)




Once you write your django app, you will need to run it on your local webserver port 8000 using following command from command prompt...python manage.py runserver, your app should run now with out errors



Now its time to use NGROK to expose your webserver to internet, use this only for development for production you need to create a secure webserver

ngrok http yourport will expose your local webserver to internet




Copy forwarding https url and paste it in your dialogflow application, your Django application is now ready to receive web-service requests.

Since we are using Django now as a webhook not PeopelSoft rest service, sky is the limit to integrate your chatbot to any third party applications to fulfill user chat requests

In part 3 I will show a working prototype of Kommunicate chat