https://github.com/noahtigner/discoverdaily

A Spotify Recommender System. Trains a Classifier on your musical tastes and recommends songs daily. Uses the Spotify API and scikit-learn for machine learning.
https://github.com/noahtigner/discoverdaily

machine-learning recommender-system scikit-learn spotify spotify-api

Last synced: about 2 months ago
JSON representation

A Spotify Recommender System. Trains a Classifier on your musical tastes and recommends songs daily. Uses the Spotify API and scikit-learn for machine learning.

• Host: GitHub
• URL: https://github.com/noahtigner/discoverdaily
• Owner: noahtigner
• Created: 2019-12-22T20:41:58.000Z (over 6 years ago)
• Default Branch: main
• Last Pushed: 2023-06-22T02:19:37.000Z (about 3 years ago)
• Last Synced: 2025-04-08T14:42:46.657Z (about 1 year ago)
• Topics: machine-learning, recommender-system, scikit-learn, spotify, spotify-api
• Language: Python
• Homepage:
• Size: 1.45 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# DiscoverDaily

A Recommender System that uses scikit-learn and the Spotify API.

Analyzes your Spotify library - including likes and dislikes - to produce music recommendations for you every day.

### Exploratory Data Analysis of my musical taste

![EDA](screenshots/eda.png)

### The Decision Tree Classifier

![Tree](screenshots/tree.png)

### Use

You will need a Spotify Client ID and Client SECRET. See: [Spotify Authorization](https://developer.spotify.com/dashboard/applications)

The application will analyze your library, as well as over 2000 songs that I have arbitrarily deemed 'bad'.

* export your SPOTIPY_USERNAME, SPOTIPY_CLIENT_ID, SPOTIPY_CLIENT_SECRET, and SPOTIPY_REDIRECT_URI

    * See [spotipy env](https://spotipy.readthedocs.io/en/latest/#authorization-code-flow)

    

* To run on a daily basis:

    * run run.sh

    

* For a single use:

    * run discoverdaily.py

    * args:

        * playlist_length: number of track recommendations to generate

### Development

Written in Python.

Dependencies: 

* sklearn for machine learning

* pandas for data management

* Spotipy, a wrapper around the Spotify API

* sqlite3 for DBM

## Author

Noah Tigner

noahzanetigner@gmail.com