Detecting and Tracking Political Abuse in Social Media

By Jacob Ratkiewicz, Michael D. Conover, Mark R. Meiss, Bruno Gonçalves, Alessandro Flammini, Filippo Menczer. 2011

645 citations recorded by the source corpus.

We study astroturf political campaigns on microblogging platforms: politically-motivated individuals and organizations that use multiple centrally-controlled accounts to create the appearance of widespread support for a candidate or opinion. We describe a machine learning framework that combines topological, content-based and crowdsourced features of information diffusion networks on Twitter to detect the early stages of viral spreading of political misinformation. We present promising preliminary results with better than 96% accuracy in the detection of astroturf content in the run-up to the 2010 U.S. midterm elections.

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

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

How algorithmic popularity bias hinders or promotes quality (2017, background): “Second, engagement and popularity metrics are subject to manipulation, for example by fake reviews, social bots, and astroturf [24, 8].”
Connotation Frames of Power and Agency in Modern Films (2017, background): “There is much prior research focus on bias in social media (Garcia et al., 2014; Prabhakaran et al., 2014; Ratkiewicz et al., 2011; Yano et al., 2010; Srivastava and Sahami, 2009), complementing our investigation on movies.”
The power of prediction with social media (2013, background): “…in social media contents is far from easy; and, of course, some users are trying to “cheat” by either spreading misinformation (e.g., Metaxas and Mustafaraj, 2010 and 2012) or by producing an abnormally high volume of conversation by means of automated accounts (e.g., Ratkiewicz et al., 2011).”
Dissecting a Social Botnet: Growth, Content and Influence in Twitter (2015, background): “Much work has focused on automatic detection mechanisms [3, 15] but little has been done to understand the behaviors and characteristics of social botnets in social computing platforms.”
Community Interaction and Conflict on the Web (2018, background): “forms has focused primarily on controversies [3, 13, 20, 21, 45, 52] and anti-social behavior, in the form of trolling [11, 12, 42], sockpuppetry [39], harassment and cyberbullying [8, 23, 32, 35, 55, 66, 79], vandalism [43], hate speech [9, 15, 51, 56, 72], and others [17, 24, 40, 41, 44, 46, 54, 6”
Dynamical classes of collective attention in twitter (2011, background): “…including its network topology [18, 19, 20], the relations and types of messages between users [21, 22], the internal information propagation [23, 24, 25], the credibility of information [26, 27], and even its potential as an indicator of the state of mind of a population [28, 29, 30, 31, 32, 25].”
Hoaxy: A Platform for Tracking Online Misinformation (2016, background): “Beginning with the detection of simple instances of political abuse like astroturﬁng [31], researchers noted the need for automated tools for monitoring such platforms.”
Predicting political preference of Twitter users (2013, methodology): “analysis systems that could, in principle, use trained models to predict user preferences ‘‘on the fly’’ (Ratkiewicz et al. 2011; Wang et al. 2012).”
Prominent Features of Rumor Propagation in Online Social Media (2013, background): “We used the Twitter data reported in [4], which comprise the profile information of 54 million users, the 1.9 billion follow links between them, and all of the 1.7 billion public tweets posted over the course of three and a half years since the launch of Twitter in March 2006.”
The spreading of misinformation online (2016, background): “To counteract this trend, algorithmic-driven solutions have been proposed (24 – 29), e.g., Google (30) is developing a trustworthiness score to rank the results of queries.”
Tampering with Twitter’s Sample API (2018, background): “Detection of bots can use content of Tweets [60] or profile information, e.”
Improving short text classification through global augmentation methods (2019, background): “n classication models that do not necessarily have a large amount of labelled data. Labelling of data comes at a cost. Whether it is for identifying fake news [17], understanding political phenomena [27], feedback on government services [29], or better coordination during emergencies [14], getting l”
No . 2017 . 5 Computational Propaganda in the United States of America : Manufacturing Consensus Online (2017, methodology): “There is evidence that US political actors have previously used bots in attempts to manipulate public opinion (Ratkiewicz et al., 2011a; Metaxas & Mustafaraj, 2012).”
Twitter as Data (2018, background): “) Astroturf political campaigns commonly use authentic looking accounts controlled by a political operation to share one or a few links, creating the appearance of grassroots concern where none exists (Mustafaraj and Metaxas, 2010; Ratkiewicz et al., 2011).”
Arming the public with artificial intelligence to counter social bots (2019, background): “During the 2010 U.S. midterm elections, primitive social bots were found to support some candidates and attack their opponents (Metaxas and Mustafaraj, 2012), injecting thousands of tweets pointing to websites with fake news (Ratkiewicz et al., 2011).”