Detecting Spam in a Twitter Network

By Sarita Yardi Schoenebeck, Daniel M. Romero, Grant Robert Schoenebeck, D. boyd. First Monday, 2009

385 citations recorded by the source corpus.

Spam becomes a problem as soon as an online communication medium becomes popular. Twitter’s behavioral and structural properties make it a fertile breeding ground for spammers to proliferate. In this article we examine spam around a one-time Twitter meme—“robotpickuplines”. We show the existence of structural network differences between spam accounts and legitimate users. We conclude by highlighting challenges in disambiguating spammers from legitimate users.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/063d8430ecd308265c4f6584d6e6fb8985eaa8a0

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

Citations in the corpus

Confounds and Consequences in Geotagged Twitter Data (2015, methodology): “To eliminate spam and automated accounts (Yardi et al., 2009), we removed tweets containing URLs, user accounts with more than 1000 followers or followees, accounts which have tweeted more than 5000 messages at the time of data collection, and the top 10% of accounts based on number of messages in…”
Detecting spammers on social networks (2010, background): “Yardi et al. [18] ran an experiment on Twitter spam.”
Detecting Spammers on Twitter (2010, background): “[32] studied the behavior of a small group of spammers, finding that they exhibit very different behavior from non-spammers in terms of posting tweets, replying tweets, followers, and followees.”
The Paradigm-Shift of Social Spambots: Evidence, Theories, and Tools for the Arms Race (2017, background): “CL8, but to no avail. To achieve eective detection results, instead of the MCL, we 8MCL admits 2 fundamental parameters: in ation and expansion. established work emerging trends Yardi, Romero et al. [45] 2009 Beutel, Faloutsos, et al. [5,18,21,22] 2013-16 Benevenuto et al. [3,4,17] 2009-12 Cao et al”
Spam Detection on Twitter Using Traditional Classifiers (2011, background): “In [13], the authors created a popular hashtag on Twitter and observed how spammers started to use it in their messages.”
Fake News Identification on Twitter with Hybrid CNN and RNN Models (2018, neutral): “Yardi et al developed three feature types for spam detection on Twitter; which includes searches for URLs, matching username patterns, and detection of keywords from supposedly spam messages[18].”
Uncovering social network sybils in the wild (2011, background): “Other studies have leveraged Bayesian ﬁlters and SVMs to identify spammers on Twitter [2,17,20] and Facebook [14].”
Digital Footprints: Opportunities and Challenges for Online Social Research (2014, background): “…to tell who is a human; the growing incidence of spam accounts (fake Twitter accounts created to send marketing or other unwelcome messages) is worrisome, and despite progress in spam-detection methods (Yardi et al. 2010), spammers manage to circumvent these methods and keep the “arms race” going.”
A decade of social bot detection (2020, background): “The first work that specifically addressed the detection of automated accounts in online social networks dates back to January 2010 [37].”
On the hardness of evading combinations of linear classifiers (2013, background): “In a growing number of adversarial domains, including many kinds of spam and fraud, machine learning is being used to detect malicious behavior [5, 6, 8, 18, 17, 14, 13, 3].”
SybilSCAR: Sybil detection in online social networks via local rule based propagation (2017, neutral): “, IP address, behaviors, and content) [22], [23], [24], [25], [26], [27].”
Combating Attacks and Abuse in Large Online Communities (2016, methodology): “To begin with, we examine the possibility to buildmachine learning classifiers to detect workers who perform crowdturfing tasks (on Weibo, Chinese Twitter).”
Predicting the Political Alignment of Twitter Users (2011, background): “The detection of spam accounts and content is an active area of research [14]–[16].”