Online Human-Bot Interactions: Detection, Estimation, and Characterization

By Onur Varol, Emilio Ferrara, Clayton A. Davis, Filippo Menczer, Alessandro Flammini. International Conference on Web and Social Media, 2017

1,106 citations recorded by the source corpus.

Increasing evidence suggests that a growing amount of social media content is generated by autonomous entities known as social bots. In this work we present a framework to detect such entities on Twitter. We leverage more than a thousand features extracted from public data and meta-data about users: friends, tweet content and sentiment, network patterns, and activity time series. We benchmark the classification framework by using a publicly available dataset of Twitter bots. This training data is enriched by a manually annotated collection of active Twitter users that include both humans and bots of varying sophistication. Our models yield high accuracy and agreement with each other and can detect bots of different nature. Our estimates suggest that between 9% and 15% of active Twitter accounts are bots. Characterizing ties among accounts, we observe that simple bots tend to interact with bots that exhibit more human-like behaviors. Analysis of content flows reveals retweet and mention strategies adopted by bots to interact with different target groups. Using clustering analysis, we characterize several subclasses of accounts, including spammers, self promoters, and accounts that post content from connected applications.

Subfields: Artificial Intelligence, Information Systems, Sociology and Political Science

View Paper: https://www.semanticscholar.org/paper/6a025e1e4d7269b20dadc91e14f28a93124fc145

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

In Sociology and Political Science

Citations in the corpus

A Survey of Fake News (2020, background): “One recent study— that classifies accounts based on observable features such as sharing behavior, the number of social ties, and linguistic features—estimates that between 9% and 15% of active Twitter accounts are bots [Varol et al. 2017].”
Contrast Pattern-Based Classification for Bot Detection on Twitter (2019, background): “There are currently about 48 million bot accounts in Twitter, amounting approximately to 15% of all active client accounts [8].”
Fake news game confers psychological resistance against online misinformation (2019, neutral): “%) are bots (Varol et al., 2017).”
Analyzing the Digital Traces of Political Manipulation: The 2016 Russian Interference Twitter Campaign (2018, background): “…at scale, by cloning or emulating the activity of human users while operating at much higher pace (e.g., automatically producing content following a scripted agenda) [26, 34, 38, 54] – however, it should be noted that bots have been also used, in some instances, for positive interventions [40, 47].”
Scalable and Generalizable Social Bot Detection through Data Selection (2019, methodology): “To obtain the varol-icwsm dataset, the authors manually labeled accounts sampled from different Botometer score deciles (Varol et al. 2017).”
Fake news on Twitter during the 2016 U.S. presidential election (2019, methodology): “In order to further guarantee that the sample did not include bots, we identified accounts judged by BorOrNot as bot-like (35).”
Arming the public with artificial intelligence to counter social bots (2019, methodology): “A recent report 2 revealed that reality television stars, professional athletes, comedians, TED speakers, pastors, and fashion models all purchased fake followers on Twitter at least once.”
Predictive Analysis on Twitter: Techniques and Applications (2018, methodology): “Sentiment has also been used in spam detection works [119, 120] as a feature to detect bots on Twitter.”
Deepfakes generation and detection: state-of-the-art, open challenges, countermeasures, and way forward (2021, methodology): “A study published in [22, 23] concludes that during the US election campaign-2016, bots were employed to generate one-fifth of the tweets during the last month of the campaign.”
The future of social media in marketing (2019, background): “For example, experts estimate that up to 15% of active Twitter accounts are bots (Varol et al.”
Characterizing the 2016 Russian IRA influence campaign (2018, background): “The number of users that have bot scores above 0.5, and can therefore be safely considered bots according to prior work Varol et al. (2017b), stands at 8,424 accounts.”
An Overview of Catastrophic AI Risks (2023, background): “While there are already many social media bots [19], some of which exist to spread disinformation, historically they have been run by humans or primitive text generators.”
A decade of social bot detection (2020, background): “was estimated to be in the region of 15% of all active Twitter accounts in 2017 [31], and 11% of all Facebook accounts in 2019 [38] – a considerable share indeed.”