logo Netzschleuder network catalog, repository, and centrifuge

discogs_label — Discogs label-genre and label-style affiliations (2015)

Description

Two bipartite networks of record labels and the genres or styles of their releases, from the Discogs database. Discogs (discogs.com) is a user-built online database of music recordings, whose contributors catalog releases (albums, singles, compilations, etc.) with their labels, and classify each release into one or more of 15 broad genres (e.g. Rock, Electronic, Jazz) and, optionally, into finer styles (e.g. Deep House, Free Jazz, Punk); its content is published as monthly XML dumps. These networks are built from the releases dump of 1 October 2015, which contains the 6,379,659 releases accepted in the database at that time (drafts and the few rejected or deleted releases of the dump are excluded). Nodes are labels and genres (genre) or styles (style), and an undirected edge connects a label to a genre or style if the label has at least one release of that genre or style, weighted by the number of such releases. When a release has several labels (co-releases, sublabels, distributors listed as labels), each of them is connected to the genres and styles of the release.

  • genre: labels and genres.
  • style: labels and styles.

The dump of 2015 does not contain label ids, and labels are therefore identified by their names, which are unique in Discogs. Discogs uses placeholder labels such as "Not On Label" (with variants for self-released records), which are kept as nodes. Labels without any release with a genre (or style) are not included in the corresponding network.

The networks have the following properties:

  • (Node) is_label: whether the node is a label (otherwise it is a genre or style).
  • (Node) name: the name of the label, genre or style.
  • (Edge) count: the number of releases of that label with that genre or style.
  • (Edge) first_year and last_year: the earliest and latest release years among them (-1 if none of them has a known year; the years are as entered in Discogs, and a few are evidently mistyped).
Tags
Informational Relatedness Affiliation Weighted Metadata
Added
2020-08-11
Citation
Upstream URL OK
https://data.discogs.com/
Upstream license
Creative Commons CC0 1.0 Universal (Public Domain Dedication), https://creativecommons.org/publicdomain/zero/1.0/
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
genre 497,332 913,948 3.68 485.15 434.84 4.47 -0.55 0.00 4 1.00 Undirected Bipartite is_label name count first_year last_year 12.4 MiB 19.4 MiB 18.9 MiB 17.2 MiB
style 434,030 1,862,824 8.58 264.58 453.28 5.49 -0.41 0.00 6 1.00 Undirected Bipartite is_label name count first_year last_year 13.3 MiB 23.0 MiB 22.3 MiB 19.9 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.