About

A basic chatbot using Rasa for accessing info about languages from the World Atlas of Languages data in German. This chatbot is built as part of the NLP course for Plaksha University taught by Monojit Choudhary.

Installation

• Windows Installation Instructions
• Linux Installation Instructions
• MacOS Installation Instructions

How to run Locally

Note: You need to first go to the complete_version folder in your terminal and then run the following commands.

1. You can train the models by running:
   rasa train
   This will train the Rasa NLU and Core models and store them inside inside the models/ folder of your project directory.

2. In a new terminal (or anaconda prompt if on windows) start the server for the custom action by running:
   rasa run actions
   This will start the server for emulating the custom action.

3. Talk to the assistant by running:
   rasa shell
   This will load the assistant in your terminal for you to chat.

Deploying to Slack

1. Go to your Slack app's settings page and use the Bot User OAuth Access Token and Signing Secret:

And add this in the credentials.yml file:

slack:
  slack_token: "Bot User OAuth Access Token"
  slack_signing_secret: "Signing Secret"
  slack_channel: 

2. Start the action server by typing the following command in terminal:

rasa run actions

3. Start the rasa server in another terminal window:

rasa run -m models --enable-api --cors "*" --debug

This will start the server at port 5005.

6. Now you have to expose this port to the world by using ngrok, open another terminal and type:

ngrok http 5005

7. Take the above url and paste it into the Events Subscription page of your slack app in the following format:

your_url_here/webhooks/slack/webhook

And you should now be able to talk to your chatbot in Slack!

License

GPL-V3