A sub-sample of the Spotify song data originally collected by Kaylin Pavlik (kaylinquest) and distributed through the R for Data Science TidyTuesday project.
spotify
A data frame with 350 songs (or tracks) and 23 variables:
unique song identifier
song name
song artist
song popularity from 0 (low) to 100 (high)
id of the album on which the song appears
name of the album on which the song appears
when the album was released
Spotify playlist on which the song appears
unique playlist identifier
genre of the playlist
subgenre of the playlist
a score from 0 (not danceable) to 100 (danceable) based on features such as tempo, rhythm, etc.
a score from 0 (low energy) to 100 (high energy) based on features such as loudness, timbre, entropy, etc.
song key
song loudness (dB)
0 (minor key) or 1 (major key)
a score from 0 (non-speechy tracks) to 100 (speechy tracks)
a score from 0 (not acoustic) to 100 (very acoustic)
a score from 0 (not instrumental) to 100 (very instrumental)
a score from 0 (no live audience presence on the song) to 100 (strong live audience presence on the song)
a score from 0 (the song is more negative, sad, angry) to 100 (the song is more positive, happy, euphoric)
song tempo (beats per minute)
song duration (ms)