We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Computer Science > Social and Information Networks

Title: A study on Channel Popularity in Twitch

Abstract: In the past few decades, there has been an increasing need for Internet users to host real time events online and to share their experiences with live, interactive audiences. Online streaming services like Twitch have attracted millions of users to stream and to spectate. There have been few studies about the prediction of streamers' popularity on Twitch. In this paper, we look at potential factors that can contribute to the popularity of streamers. Streamer data was collected through consistent tracking using Twitch's API during a 4 weeks period. Each user's streaming information such as the number of current viewers and followers, the genre of the stream etc., were collected. From the results, we found that the frequency of streaming sessions, the types of content and the length of the streams are major factors in determining how much viewers and subscribers streamers can gain during sessions.
Subjects: Social and Information Networks (cs.SI); Machine Learning (cs.LG)
ACM classes: H.3.3
Cite as: arXiv:2111.05939 [cs.SI]
  (or arXiv:2111.05939v1 [cs.SI] for this version)

Submission history

From: Ha Le [view email]
[v1] Wed, 10 Nov 2021 20:55:20 GMT (2932kb,D)

Link back to: arXiv, form interface, contact.