twitch_social — Twitch mutual follower networks (2018)

Description

Social networks of users of the live streaming platform Twitch, collected from its public API in 2018 by B. Rozemberczki and collaborators for benchmarks of node embedding and classification methods. Nodes are Twitch users, and an undirected edge connects two users who follow each other (a mutual follower relationship). The user identifiers were replaced by consecutive indices. There are two data sets:

Language networks (MUSAE, collected in May 2018): six networks of users who stream in a given language, DE (German), ENGB (English), ES (Spanish), FR (French), PTBR (Portuguese) and RU (Russian). Each node has a list of binary features, shared by all six networks, which the authors extracted from the games played and liked, the location and the streaming habits of the user; only the indices of the features are given, not their meaning. The supervised task proposed with the data is to predict whether a streamer uses explicit language (mature). In FR, two users appear twice in the node table (with the same values, except for a difference of 1 in the views of one of them); the first row is used.
twitch_gamers (collected in spring 2018): a larger network of 168,114 users of all languages, which forms a single connected component. It has no feature lists, but more user attributes.

The networks have the following properties:

(Node) twitch_id (language networks): the numeric Twitch identifier of the user.
(Node) days (language networks): the age of the account in days (as given in the data).
(Node) mature: whether the user streams explicit (mature) content.
(Node) views: the number of views.
(Node) partner (language networks): whether the user is a Twitch partner.
(Node) features (language networks): the indices of the user's binary features (as listed in the data, in which a few lists contain repeated indices).
(Node) life_time (twitch_gamers): the age of the account in days.
(Node) created_at, updated_at (twitch_gamers): the dates of creation and of the last update of the account, as Unix time (at midnight UTC).
(Node) dead_account (twitch_gamers): whether the account is inactive.
(Node) language (twitch_gamers): the broadcaster language (a two-letter code, or "OTHER").
(Node) affiliate (twitch_gamers): whether the user is a Twitch affiliate.

Tags

Social Online Unweighted Metadata

Added

2026-10-27

Citation

B. Rozemberczki, C. Allen, and R. Sarkar, "Multi-scale attributed node embedding." Journal of Complex Networks 9(2), cnab014 (2021). 10.1093/comnet/cnab014 Sci-Hub
B. Rozemberczki and R. Sarkar, "Twitch Gamers: a Dataset for Evaluating Proximity Preserving and Structural Role-based Node Embeddings." arXiv:2101.03091 (2021). 10.48550/arXiv.2101.03091 Sci-Hub

Upstream URL OK

https://snap.stanford.edu/data/twitch-social-networks.html
https://snap.stanford.edu/data/twitch_gamers.html
https://github.com/benedekrozemberczki/datasets

Upstream license

MIT License, https://github.com/benedekrozemberczki/datasets/blob/3b19135105432cbca65765c1113b91c395ed9d44/LICENSE

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	gt	GraphML	GML	csv
DE	9,498	153,138	32.25	84.80	145.59	5.01	-0.12	0.05	7	1.00	Undirected	Unipartite	twitch_id days mature views partner features	663 KiB	1016 KiB	1021 KiB	1.0 MiB
ENGB	7,126	35,324	9.91	22.19	39.48	8.74	-0.12	0.04	10	1.00	Undirected	Unipartite	twitch_id days mature views partner features	371 KiB	510 KiB	500 KiB	560 KiB
ES	4,648	59,382	25.55	49.39	87.41	8.00	-0.19	0.08	9	1.00	Undirected	Unipartite	twitch_id days mature views partner features	291 KiB	446 KiB	427 KiB	453 KiB
FR	6,549	112,666	34.41	77.49	126.44	4.02	-0.18	0.05	7	1.00	Undirected	Unipartite	twitch_id days mature views partner features	465 KiB	730 KiB	722 KiB	745 KiB
PTBR	1,912	31,299	32.74	55.84	96.90	4.41	-0.23	0.13	7	1.00	Undirected	Unipartite	twitch_id days mature views partner features	129 KiB	208 KiB	199 KiB	198 KiB
RU	4,385	37,304	17.01	41.68	72.62	7.73	-0.18	0.05	9	1.00	Undirected	Unipartite	twitch_id days mature views partner features	251 KiB	366 KiB	353 KiB	382 KiB
twitch_gamers	168,114	6,797,557	80.87	314.16	572.49	8.73	-0.09	0.02	8	1.00	Undirected	Unipartite	views mature life_time created_at updated_at dead_account language affiliate	16.6 MiB	28.9 MiB	29.0 MiB	27.6 MiB