sousablde / movichtr-with-rasa

Rasa based conversational bot on the thematics of movie related content (with the help base of TMDB as source material)

MoviCHTR

A conversational bot on movie thematics

• the bot finds:
  • movies a director is known for
  • movies and actor is known for
  • the release date of a movie if the movie title is provided
  • the overview/plot of a movie if a title is provided
  • searches for the rating of a movie based on movie title
  • Finds popular movies by year
  • engages in basic chit-chat

The project goal was to:

• Understanding the Rasa framework
• Installing the Rasa framework
• Experience with manipulating language based data

Software used

Everything is programmed in python, to follow this project you need the following

• Python <=3.6.8
• Jupyter Notebook
• Ngrok
• Slack

The source tools used within Jupyter notebook are

• rasa nlu
• rasa core
• rasa x

Additional support from:

• text editor
• web browser
• terminal

File overview:

data/core/ - contains stories

data/nlu - contains example NLU training data

demo - contains custom action/api code

domain.yml - the domain file

config.yml - the Rasa config file

events. - files related to rasa x usage

Testing

• Within jupyter notebooks comment out slack related code from within the loading assistant definition, engage with load_assistant()
• outside of notebook launch rasa x for a webbrowser based testing and interactive learning
• from slack: request server start and engage with @movichtr

More extended step by step information and visual tutorials provided within the notebook.