spotify-recommender-system

Spotify Recommender System was the final project for my master's dissertation. It consists in a platform that uses Inverse Reinforcement Learning to analyse and infer user preferences on a streaming platform, in this case, Spotify. It later recommends new content to users according to their musical taste. It holds a client-side in charge of communicating with the Spotify's API using React and a sever-side application built on Python that treats the acquired information.

Instructions

Setup

Access Spotify for Developers and create a new app. That should give you a Client ID and a Client Secret.

Inside the client/ directory, create a .env file and add those two tokens accordingly:

• REACT_APP_CLIENT_ID = ...
• REACT_APP_CLIENT_SECRET = ...

Run it

Create 2 terminal instances.

On the first one:

1. Change to the server directory: cd server/
2. Install the dependencies:
   • pip install Flask
   • pip install flask-cors
   • pip install pulp
3. python3 server.py

On the second:

1. Change to the client directory: cd client/
2. Install the dependecies: npm install
3. Start the app: npm start

To test the application you only need to use the client app, by accessing http://localhost:3000/.