Spotify Charts weekly top 200 songs between 2017~2021 in global dataset & api extraction codes
spotify weekly top 200 songs between 2017-2021 in global.
74.661 rows
40 columns
2.986.440 items
10 data types (char, tinyint, varchar, bool, date, smallint, decimal, blob, float, int)
columns.
track_id(char[22]) : spotify id for the track
track_name(varchar[666]) : name of the track
track_popularity(double[3, 2]) : popularity of the track calculated by spotify
track_number(u-tinyint) : track’s index relative to its album
album_id(char[22]) : spotify id for the album that the track is from
album_name(varchar[666]) : name of the album that the track is from
album_img(blob) : link to the cover image of album that the track is from
album_type(varchar[10]) : type of the album (eg. single, album)
album_label(varchar[666]) : track’s record label
album_track_number(u-tinyint) : number of the tracks in the album that the track is from
album_popularity(double[3,2]) : popularity of the album calculated by spotify
artist_num(u-tinyint) : number of artists in the track
artist_names(varchar[666]) : names of all artists who participated in the track (separated by comma)
artist_id(char[22]) : spotify artist id for the artist_individual
artist_name(varchar[666]) : one of the artists who participated in the track (tracks with multiple artists are split into separate rows for each artist)
artist_img(char[40]) : link to the artist_individual’s image
artist_followers(u-int) : follower amount of artist
artist_popularity(decimal[3,2]) : popularity of the artist calculated by spotify
artist_genres(varchar[666]) : artist’s genres
rank(u-tinyint) : ranking of the track on the chart
week(date) : end of week the track was in charts as date format
streams(u-int) : number of streams in that week
collab(bool) : if the participation of the track is multiple or not (0 if there is only one artist, else 1)
explicit(bool) : explicit situation of the track
release_date(date) : release date of the album (thus track)
danceability(double[4, 3]), energy(double[4, 3]), key(tinyint), mode(bool), time_signature(u-tinyint), loudness(float), speechiness(decimal[5,4]), acousticness(float), instrumentalness(float), liveness(decimal[5,4]), valence(decimal[5,4]), tempo(float), duration(u-int) : Spotify API Reference
pivot(bool) : when multiple artists are split into separate rows, this value takes 0 for the first artist and 1 for the rest
spotify api docs~ developers spotify
pivot~ bartomiejczyewski's spotify top 200
idea~ yelexa's spotify top 200
kaggle~ my kaggle page for the dataset