MoodSick

A repository housing our implementation for the CIS 600 (Social Media and Data Mining) project

Environment set-up

We are using docker to manage our environment. This will help us ensure that we will all be using the same environment and will not run into any issues with dependencies.

Install Docker

• Windows
• Mac
• Linux

Build the docker image

• Clone the repository
• Navigate to the docker directory:

  cd docker
  

• Build the docker environment:

  docker-compose build
  

• Run the docker environment:

  docker-compose up
  

Docker network

• The docker-compose file creates a network called moodSickNetwork
• This network is used to connect all of our services together
  • Frontend : http://localhost:3000
  • Middleware : http://localhost:8080
  • Ml model : http://localhost:5000
  • Database : http://localhost:27018
• The below diagram shows the architecture of our application

Environment Files

• The docker-compose file uses environment files to set up the environment variables
• There are two environment files
  • docker/middleware/.env : This file contains the environment variables for the Spotify API

      SPOTIFY_CLIENT_ID,
      SPOTIFY_CLIENT_SECRET,
      SPOTIFY_USER_ID
  

  • docker/ml/.env : This file contains the environment variables ATLAS_URI which has the vectors used by the ML model

Exploration

If you'd like to do some data exploration, open the notebook located at the following location:

• Initial exploration: Initial Exploration
• Mood Genere Dynamics: Mood Genre Dynamics

Contributors

Acknowledgements

We'd like to thank the authors of the GZTAN and the Spotify datasets.