lastfm — Last.fm listening histories (2009)

Description

Bipartite networks of the users of the music website Last.fm and the artists and songs they listened to, from the complete listening histories (until 5 May 2009) of 992 users, collected by Òscar Celma through the Last.fm API (method user.getRecentTracks). Each edge is a listening event ("scrobble"): an edge connects a user to an artist (network band) or to a song (network song) each time the user listened to a song of the artist, or to the song, with the time of the event, so that the networks are multigraphs with 19,150,868 edges each. All users of the data set's user profiles are included.

The artists and songs are identified by their MusicBrainz identifiers (MBID), or, for those without one, by their names as given: an artist without an MBID by its name, and a song without an MBID by its name together with its artist (so that the same artist or song may appear under different spellings, and several artists with the same name are merged; different songs with the same title are only merged if they are by the same artist without an MBID). An artist or song with an MBID that appears with several names is named after its first occurrence.

The networks have the following properties:

(Node) name: the identifier of the user (e.g. user_000639), or the name of the artist or song.
(Node) is_user: whether the node is a user (otherwise an artist or a song).
(Node) mbid: the MusicBrainz identifier of the artist or song (empty for users and for artists and songs without one).
(Node) artist, artist_mbid: the name and MusicBrainz identifier of the artist of the song (network song only; empty for users).
(Node) gender: the gender of the user (m, f, or empty if not given; empty for artists and songs).
(Node) age: the age of the user (-1 if not given, and for artists and songs).
(Node) country: the country of the user (empty if not given, and for artists and songs).
(Node) signup: the date of registration of the user, in Unix time (-1 if not given, and for artists and songs).
(Edge) time: the time of the listening event, in Unix time.

Tags

Social Online Preferences Bipartite Unweighted Multigraph Timestamps Metadata

Added

2020-08-11

Citation

Ò. Celma, "Music Recommendation and Discovery: The Long Tail, Long Fail, and Long Play in the Digital Music Space." Springer (2010). 10.1007/978-3-642-13287-2 Sci-Hub

Upstream URL OK

https://www.upf.edu/web/mtg/lastfm1k
https://zenodo.org/records/6090214

Upstream license

Non-commercial use (distributed with permission of Last.fm), https://zenodo.org/records/6090214

Networks

Tip: click on a column header to sort the list, or hover over it for a description.

Column legend

$\left<k\right>$: Mean degree
$\sigma_k$: Standard deviation of the degrees
$\lambda_h$: Largest eigenvalue of the Hashimoto (non-backtracking) matrix
$\tau$: Random walk mixing time
$r$: Degree assortativity
$c$: Global clustering coefficient
$\oslash$: (Pseudo-) diameter
$S$: Relative size of the largest component
NPs: Node properties
EPs: Edge properties

Name	Nodes	Edges	$\left<k\right>$	$\sigma_k$	$\lambda_h$	$\tau$	$r$	$c$	$\oslash$	$S$	Kind	Mode	NPs	EPs	gt	GraphML	GML	csv
band	175,069	19,150,868	218.78	2524.08	359.89	12.99	-0.27	0.00	6	1.00	Undirected	Bipartite		weight time	72.8 MiB	101.9 MiB	97.0 MiB	95.3 MiB
song	1,085,612	19,150,868	35.28	915.72	589.26	11.34	-0.44	0.00	6	1.00	Undirected	Bipartite		weight time	97.6 MiB	141.5 MiB	137.6 MiB	135.4 MiB

Ridiculograms^*

^* These are automatically generated force-directed visualizations, and can be quite meaningless for networks both big and small. They should not be taken seriously as sources of scientific insight. See here for a discussion.