Source-linked AI summary
What is Gab? A Bastion of Free Speech or an Alt-Right Echo Chamber?
Savvas Zannettou, Barry Bradlyn, Emiliano De Cristofaro, Haewoon Kwak, Michael Sirivianos, Gianluca Stringhini, Jeremy Blackburn
TL;DR
Fringe online communities can influence the Web’s information ecosystem, yet their evolution and activities remain insufficiently characterized. This paper gives the first characterization of Gab using a large longitudinal crawl, finding that it is a news- and world-events-oriented platform attracting alt-right users, conspiracy theorists, and trolls, with hate speech more prevalent than on Twitter but less prevalent than on 4chan’s /pol/.
Problem
Limited evidence about how emerging fringe communities evolve and what activities they attract motivates monitoring their impact on the Web’s information ecosystem.
Method
The authors characterize Gab by crawling 22,112,812 posts, 336,752 users, account details, and follower/following networks between August 2016 and January 2018.
Results
Gab attracts alt-right users, conspiracy theorists, and trolls, centers on news and world events, and has 5.4% of posts containing hate words—2.4 times Twitter’s rate and 2.2 times lower than /pol/’s.
Takeaways & Limitations
Gab lies between mainstream networks such as Twitter and fringe communities such as 4chan’s /pol/, including coordinated efforts to recruit millennials to the alt-right.
Abstract
from arXiv · showhide
Over the past few years, a number of new "fringe" communities, like 4chan or certain subreddits, have gained traction on the Web at a rapid pace. However, more often than not, little is known about how they evolve or what kind of activities they attract, despite recent research has shown that they influence how false information reaches mainstream communities. This motivates the need to monitor these communities and analyze their impact on the Web's information ecosystem. In August 2016, a new social network called Gab was created as an alternative to Twitter. It positions itself as putting "people and free speech first'", welcoming users banned or suspended from other social networks. In this paper, we provide, to the best of our knowledge, the first characterization of Gab. We collect and analyze 22M posts produced by 336K users between August 2016 and January 2018, finding that Gab is predominantly used for the dissemination and discussion of news and world events, and that it attracts alt-right users, conspiracy theorists, and other trolls. We also measure the prevalence of hate speech on the platform, finding it to be much higher than Twitter, but lower than 4chan's Politically Incorrect board.
1 Introduction
This paper presents the first characterization of Gab, examining its users, discussions, hate speech, and relationship to mainstream and fringe communities. Gab attracts alt-right users, conspiracy theorists, trolls, and coordinated alt-right recruitment efforts, while hate speech is more prevalent than on Twitter but less prevalent than on 4chan’s /pol/.
- Study scope: The study provides the first characterization of Gab, using 22M posts from 336K users collected between August 2016 and January 2018.The dataset covers approximately 1.5 years of activity on the platform.
- User communities: Gab attracts alt-right personalities, conspiracy theorists, and trolls migrating from or inspired by platforms such as 4chan.The authors identify a wide variety of user types, including well-known alt-right figures and accounts associated with trolling.
- Content and events: Gab is predominantly used to disseminate and discuss world events, news, and conspiracy theories, reacting strongly to white nationalism and Donald Trump.The analysis identifies these subjects as central to Gab activity and highlights strong reactions to related events.
- Hate speech: 5.4% of posts include hate words, a rate 2.4 times higher than Twitter but 2.2 times lower than 4chan’s Politically Incorrect board (/pol/).These comparisons position Gab between Twitter and 4chan’s /pol/ in prevalence of hate words.
- Recruitment: Several accounts make coordinated efforts to recruit millennials to the alt-right.The introduction identifies recruitment as a distinct activity occurring on the platform.
- Interpretation: The authors position Gab between mainstream networks such as Twitter and fringe communities such as 4chan’s /pol/.They argue that Gab’s free-speech framing serves as a shield behind which its alt-right users hide.
2 Related Work
Prior research shows that information and abusive content circulate across multiple online communities, including fringe platforms that can affect mainstream networks. This paper extends that line of work by providing the first study focused on Gab’s users, topics, and hateful content.
- Mainstream networks: Earlier studies established Twitter’s importance for assessing human behavior on the Web.Kwak et al. were among the first to study Twitter in this context.
- Cross-platform influence: Research on multiple communities found that small fringe communities on Reddit and 4chan can substantially affect mainstream communities such as Twitter.That work used Hawkes Processes to study how mainstream and alternative news propagate across platforms.
- Abusive-content detection: Bag of Communities identifies abusive content using training data from nine communities across 4chan, Reddit, Voat, and Metafilter.The approach outperformed methods relying only on in-community data.
- Fringe communities: Studies of 4chan’s /pol/ found high volumes of hate speech and organized raids targeting other Web communities.One example showed users posting large numbers of abusive comments on YouTube videos over a short period.
- Research gap: This paper is the first to study Gab’s users, discussion topics, and the extent to which users share hateful content.Its contribution applies the broader multi-community research perspective to Gab.
3 Gab
Gab combines Twitter-like broadcasting and Reddit-like voting with optional topics, platform-defined categories, permissive moderation, and user-supported monetization. The study collects posts, account details, and follower networks through API-based snowball sampling.
- Platform features: Gab launched in August 2016 and combines Twitter-like 300-character broadcasts with Reddit-like voting, reposts, quotes, replies, hashtags, and mentions.Posts are called “gabs,” and the platform also supports labeled NSFW content.
- Topics and categories: Users can assign posts to user-created topics or Gab-defined categories, but assigning either is optional and topics are moderated.Gab had 15 categories at the time described in the paper.
- Voting system: A user’s score equals received up-votes minus down-votes, and users need at least 250 points to down-vote others.The voting system determines content popularity and contributes to user-level popularity.
- Moderation: Gab’s lax moderation policy permits most content but forbids illegal pornography, terrorist promotion, threats, and doxing.Legal pornography is permitted under the stated policy.
- Monetization: Gab is ad-free and relies on donations and paid Pro accounts that provide additional features and premium content creation.Pro features include livestreams, verification, longer posts, formatting, and subscriber-only content.
- Dataset: The dataset contains 22,112,812 posts from 336,752 users collected through API-based snowball sampling between August 2016 and January 2018.The crawl gathered account details, all posts for sampled users, and follower/following connections.
4 Analysis
Gab’s user base and activity are strongly associated with alt-right, conspiracy-theory, troll, and news-oriented content, while its network exhibits heavy-tailed participation and relatively high reciprocity.
- User base: Gab’s prominent users include alt-right personalities, conspiracy theorists, news outlets, and troll accounts, with few apparent counterpoints to the alt-right.The rankings use followers, account score, and PageRank to assess reach, content appreciation, and network position.
- User base: 20% of users actively changed Gab’s default account description, providing a basis for analyzing customized descriptions.The study identifies customized descriptions by detecting quoted text followed by a dash and a name.
- User account analysis: 43% of users followed nobody, while only 4% had zero followers; after roughly 10 connections, posting and follower counts diverged into a heavy tail exceeding 77K posts.Follower/following counts relate nearly linearly to posting up to around 10 connections before substantial divergence.
- User account analysis: 29.5% of network node pairs were connected both ways, compared with 71.5% connected one way, and the authors characterize Gab as more tightly knit than Twitter.The authors associate the higher reciprocity with Gab’s concentration of users sharing an alt-right ideology.
- Posts analysis: Gab’s posts prominently used alternative news domains, including Breitbart, The Gateway Pundit, and Infowars, while The Daily Stormer ranked ahead of The Hill.YouTube appeared in almost 7% of posts and Twitter in 2%; the cited alternative-news percentages were 1.4%, 0.7%, and 0.5%, respectively.
- Posts analysis: News and Politics were Gab’s most popular categories at 15.91% and 10.30%, while topics and hashtags also centered on world events, Trump, politics, AltRight, Pizzagate, and BanIslam.Posts were assigned to categories 42% of the time and topics 12% of the time.
5 Conclusion
The paper presents the first characterization of Gab, showing that it attracts alt-right supporters, conspiracy theorists, and trolls while hosting substantial discussion of news, world events, and politics. It also finds that 5.4% of posts include hate words and that Gab reacts strongly to real-world events involving white nationalism and Donald Trump.
- Gab attracts users ranging from alt-right supporters and conspiracy theorists to trolls.
- Gab is extensively used to discuss news, world events, and politics-related topics.
- 5.4% of Gab posts include hate words.
- Gab reacts strongly to real-world events focused on white nationalism and support of Donald Trump.