/Funkify

Primary LanguageJupyter NotebookGNU General Public License v3.0GPL-3.0

Funkify

An exploration of what makes funk music dance-able:

  • Via the use of spotipy (python library), the university github team Beanie Intelligence and me created a toolkit to mine data by playlist, genre or artist.
  • Using that download playlist feature, I mined a playlist of the top 200+ songs (according to https://digitaldreamdoor.com/pages/best_rb-funk.html).
  • With that playlist and using the Spotify Analyzer danceability feature which ranges from 0 to 1 , I created danceability labels:
    • Split into the following 4 Classes:
      • 0 : [0.0, 0.5)
      • 1 : [0.5, 0.7)
      • 2 : [0.7, 0.8)
      • 3 : [0.8, 1.0]
  • The report included only contains the Funkify hypothesis of the full group coursework.