If you follow my previous blog post on “Step by Step Installation of Rasa X Open Source Conversational AI,” I have given a step-by-step installation process on Rasa X Conversational AI will up and run in 10 mins.

Today, I am going to give a walkthrough on each screen and tabs how it looks.

When you execute the last command, “rasa x,” you can see a nice interactive user interface screen up and run in the default browser on the link “http://localhost:5002/interactive”.

Rasa x Initial screen

We have below vertical menus:

Talk to your bot:

This is an interactive learning chatbot where you can start and chat with a bot and test.  The bot will give you answers based on predefined intents (Intents has patterns, questions, and answers); if any mistakes on bot chat, then you can correct those mistakes.

Rasa X - Talk to Your bot

Conversations:

All conversations between bot and users are stored in this tab.  You can review the chat messages, save them for later to review and delete chat history.

If the chatbot response

Rasa X - Talk to Your bot

NLU Inbox:

This tab will give you information on the prediction of intent and confidence score.

Rasa x - NLU Inbox

Intent Insights:

Intent Insights helps you to analyze your NLU training data whether the dataset needs to improve.  Also, it directs any potential to improve the dataset.

Rasa x - Insights

Models:

To see the active model, you need to train and activate a model for a Chatbot (when you install, you will not have a model, so you have to click Training, creating a model for you to activate it.

Note: You can make only one model active at a time.

Rasa x - Models

Training -> NLU data:

Training Data section contains sentences and classifying what intent it is.  You can create your own intents too.

Rasa X - NLU Training Data

Training -> Response:

We can configure a response to how the Chatbot wants to respond.

Rasa X - Responses

Training -> Stories:

You can see a workflow detail of stories.

Rasa X - Stories

Training -> Rules:

We can create or change our own set of rules how Chabot wants to action it.

Rasa X - Rules

Training -> Configuration:

You can do model configuration how many epochs need to run for training, threshold, min, and max ngram, etc.

Rasa X - Model Configuration

Train:

We need to train the model the first time, and it will create a new model when you retrain.

Rasa X - Train

Posts on SAP:

Why we like the SAP Business Rule Framework Plus (SAP BRF+) Recipe?

Simple wireframe design for SAP FIORI UI Chatbot

A simple wireframe design for SAP FIORI UI Chatbot – Part II

How to create your own SAP Fiori Chatbot in 10 days?

How SAP Event Management helps the discrete processing industry?