Skip to content

Analysis of playlist corpus

Gather information about the available playlists and their characteristics (distributions, basic features, representations).

This issue involves (but might not be limited to) following steps:

  • Exploratory analysis (a lot has been summarized by Spotify already but making it easier to access and more visual)
  • Feature development (Is there additional information in dataset that can be leveraged? (Playlist title, Playlist description, How often a playlist was changed, Popularity of songs / playlist)
  • Thoughts on song and playlist representation (foreshadowing encoding, latent feature models, word2vec, co-occurrence, and statistical models)
Edited by Andreas Rubin-Schwarz
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information