Image for post
Image for post
Photo by 丁亦然 on Unsplash

Rasa-X Is A Unique Approach To Continuous Chatbot Improvement

And Why Conversation Driven Development Is The Future

Introduction

Why This Often Fails: Users Want To Follow The Desire Path

Image for post
Image for post
Design versus Desire

Desire Path

Image for post
Image for post
Left: Desire Path | Right: Designed Path

Accommodation

Elements OF A Conversational Approach

From Designed Paths To Accommodation

Sharing Your Bot

Image for post
Image for post
IBM Watson Assistant Preview Link

NLU Inbox: Turn NLU Training Into An Administrative Task

Rasa X Overview: Installation & Functionality

Installation

pip3 install rasa-x --extra-index-url https://pypi.rasa.com/simple
rasa x
Image for post
Image for post
Rasa X Successfully Started

Start Talking To Your Bot

rasa train
rasa shell

Interactive Learning

Image for post
Image for post
Interactive Learning Chat Window

NLU Inbox

Image for post
Image for post
List of Functionality Within Rasa X

Creating Intents & Compound Entities

Image for post
Image for post
Create a New Intent
travel_detail
I want to travel from Berlin to Stuttgart by train tomorrow.
Image for post
Image for post
Example Utterance for Intent: travel_detail
Image for post
Image for post
The word “Berlin” is selected and tagged as Entity type “from_city”
Image for post
Image for post
Continuing of Entity Annotation
Image for post
Image for post
Creating Entity Names or Types
date_time
travel_mode
from_city
to_city
Image for post
Image for post
Adding a Second Intent Example

Conclusion

Read More Here…

Written by

NLP/NLU, Chatbots, Voice, Conversational UI/UX, CX Designer, Developer, Ubiquitous User Interfaces. www.cobusgreyling.me