DOCSLIB.ORG

Visualizing Social Bot Activity During Elections

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Visualizing Social Bot Activity During Elections Bot Electioneering Volume: Visualizing Social Bot Activity During Elections Kai-Cheng Yang Pik-Mai Hui Filippo Menczer [email protected] [email protected] cnets.indiana.edu/fil Center for Complex Networks and Center for Complex Networks and Center for Complex Networks and Systems Research, Indiana University Systems Research, Indiana University Systems Research, Indiana University Bloomington, Indiana Bloomington, Indiana Bloomington, Indiana ABSTRACT bots that aimed to effectively push their messages to the target It has been widely recognized that automated bots may have a sig- audience [1]. In particular, bots were most active in the core of nificant impact on the outcomes of national events. It is important the misinformation-sharing network [10] and effectively amplified to raise public awareness about the threat of bots on social media the spread of low-credibility content by posting it within seconds during these important events, such as the 2018 US midterm elec- and by targeting influential accounts [9]. Analogous automated tion. To this end, we deployed a web application to help the public campaigns were reported in countries around the globe [4, 11]. explore the activities of likely bots on Twitter on a daily basis. The Here we present Bot Electioneering Volume (BEV), a platform application, called Bot Electioneering Volume (BEV), reports on that visualizes the volume generated by bots and the corresponding the level of likely bot activities and visualizes the topics targeted targeted topics. BEV tracks online traffic centered around elections by them. With this paper we release our code base for the BEV framework, with the goal of facilitating future efforts to combat malicious bots on social media. KEYWORDS bot activity, Twitter, elections ACM Reference Format: Kai-Cheng Yang, Pik-Mai Hui, and Filippo Menczer. 2019. Bot Electioneer- ing Volume: Visualizing Social Bot Activity During Elections. In The Web Conference 2019, May 13–17, 2019, San Francisco, CA. ACM, New York, NY, USA, 3 pages. 1 INTRODUCTION Social bots have drawn great attention from the public recently. They are accounts on social media platforms controlled at least in part by algorithms to generate/share/retweet content and interact with human users [5]. The automated nature of social bots makes it easy to achieve scalability, with which a single person is capable to control thousands of accounts on one or more social media platforms. When needed, these social bots can work collectively to manipulate the public by promoting certain accounts or opinions. Being social animals, human users are inevitably vulnerable to arXiv:1902.02339v1 [cs.CY] 6 Feb 2019 the efforts of the social bots. Studies have shown ubiquitous social bots [12] distort online discussions, and particularly those about politics. During the 2010 US midterm election, primitive social bots were used to attack some candidates [6] and spread tweets with links to fake news websites [7]. A similar pattern emerged Figure 1: Illustration of BEV’s (a) crawler, (b) database, (c) an- in the 2016 US presidential election, only with more sophisticated alyzer, and screen shot of (d) front-end interface. The upper panel of the frontend shows the Bot Electioneering Volume Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for the past 8 days. Users can select a day of interest to ex- for profit or commercial advantage and that copies bear this notice and the full citation plore the top topics of that day. The bottom panels show a on the first page. Copyrights for components of this work owned by others than the tag cloud and entity lists for the selected day. The tag cloud author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission presents entities all together, with the size of each entity pro- and/or a fee. Request permissions from [email protected]. portional to how often it is tweeted by bots. The entity lists The Web Conference ’19, May 13–17, 2019, San Francisco, CA display hashtags, mentions, and links ranked by how often © 2019 Copyright held by the owner/author(s). Publication rights licensed to ACM. each is tweeted by bots. The Web Conference ’19, May 13–17, 2019, San Francisco, CA Yang, et al. from Twitter by feeding the streaming API with a list of selected 1e6 2.5 Number of election tweets hashtags. By incorporating the bot-detection ability of Botometer Number of unique users [3, 12, 13], BEV is able to distinguish between content generated by 2.0 likely bots and humans. The measurement of average bot activity is 1.5 then compared with random samples of tweets to produce a number Number 1.0 that quantifies electioneering activity by bots. BEV also collects 0.5 (a) content topics, including hashtags, mentions, and URLs shared by 0.0 likely bots, and reports on their relative volumes. BEV (botometer.iuni.iu.edu/bev) monitored public tweets about 1.6 1.5 the 2018 US midterm elections between October 22, 2018 and Dec 1.4 30, 2018. During the collection period, BEV drew over 3,000 visits. 1.3 Election tweets Random sample An archive of the data from Oct to Dec 2018 remains publicly 1.2 available for retrospective inspections. We plan to activate BEV 1.1 (b) Average bot score 1.0 again for future elections. Our goal is to raise public awareness of bot activities and their impact during elections in the past, and more importantly those in the future. 60% 50% 2 SYSTEM DESIGN 40% BEV The BEV system contains 4 major parts: a crawler, a database, an 30% (c) analyzer, and a front-end interface, as illustrated in Figure 1. 20% The crawler is in charge of tracking Twitter’s filtering API for public election-related tweets, querying Twitter’s Spritzer API for 23 Oct 201802 Nov 201812 Nov 201822 Nov 201802 Dec 201812 Dec 201822 Dec 201801 Jan 2019 random samples of public tweets, and fetching bot scores. Crawled data is stored in the database. The analyzer then extracts the re- quired information and generates the statistics for the visualization Figure 2: (a) Number of election tweets and unique users for at the application frontend. The frontend has three major parts: the each day. (b) Average bot scores for election tweets and ran- Bot Electioneering Volume timeline, a tag cloud, and entity lists. dom sample. (c) BEV timeline. The 2018 US midterm elec- The Bot Electioneering Volume measures the activity of likely bots, tions day, Nov 6, is highlighted by a vertical dashed line. while the tag could and entity lists display the topics that are most tweeted by likely bots. By clicking on the links in an entity list, 2.2 Bot identification users are directed to Hoaxy (hoaxy.iuni.iu.edu) [8], where they can explore more in-depth visualizations of the influence of bots around BEV uses Botometer (botometer.iuni.iu.edu) [3, 12, 13] to obtain bot the entities on Twitter in the recent minutes/hours/days. scores for Twitter accounts involved in election-related discourse. The data collection runs in a streaming fashion, but fetching bot Botometer is a supervised machine learning algorithm that consid- scores and analyzing the data take time. The front-end interface is ers more than a thousand features about an account and its activity updated every 4 hours to reflect newly incoming data. to estimate the likelihood that the account is automated. We con- sider accounts with bot score above 4 (on a 5-point scale) as social 2.1 Data collection bots. This is a fairly conservative threshold choice, corresponding to a posterior probability of automation near 50% [13] based on a The collection of election-related tweets is crucial to our applica- 15% prior probability [12]. tion. Our collection process starts with a set of election-related hashtags that are tracked using the Twitter filtering API. We seeded 2.3 Bot activity measurement our set of hashtags with several widely-used political hashtags in- To measure the bot electioneering volume, we first take daily av- cluding #2018midterms, #maga, and #bluewave. We then repeatedly erages of the bot scores of accounts generating political tweets, expanded the set with co-occuring hashtags [2], resulting in a set of weighted by their tweet frequencies. Considering that spamming 110 hashtags. From this set we manually removed 6 hashtags that the same message is a common strategy for bots, the weighted are general and irrelevant to election. We also added the hashtags average better highlights the amount of bot-generated content and for each US state’s Senate race: #casen, #nysen, and so on. The full their potential influence. To obrain a baseline of bot activity, we list of hashtags can be found in the FAQ page of the BEV website. produce the same weighted average for random tweets from Twit- This methodology allows our system to collect most tweets with ter’s Spritzer API, with a rate of 1,000 tweets per hour. The daily newly emerging election hashtags, because it is likely that these average bot scores for election and random tweets are shown in tweets contain some of the hashtags in our list as well. Figure 2(b). Twitter’s free filtering API offers at most 1% of Twitter’s traffic. The Bot Electioneering Volume is defined by the relative differ- We estimated that the traffic captured by our method is about 0.3– ence between the two averages: 0.5% of Twitter’s complete traffic (see Figure 2(a)).
Load more

Recommended publications
  • Arxiv:1805.10105V1 [Cs.SI] 25 May 2018
    Effects of Social Bots in the Iran-Debate on Twitter Andree Thieltges, Orestis Papakyriakopoulos, Juan Carlos Medina Serrano, Simon Hegelich Bavarian School of Public Policy Technical University of Munich Abstract Ahmadinejad caused nationwide unrests and protests. As these protests grew, the Iranian government shut down for- 2018 started with massive protests in Iran, bringing back the eign news coverage and restricted the use of cell phones, impressions of the so called “Arab Spring” and it’s revolution- text-messaging and internet access. Nevertheless, Twitter ary impact for the Maghreb states, Syria and Egypt. Many reports and scientific examinations considered online social and Facebook “became vital tools to relay news and infor- mation on anti-government protest to the people inside and networks (OSN’s) such as Twitter or Facebook to play a criti- 1 cal role in the opinion making of people behind those protests. outside Iran” . While Ayatollah Khameini characterized the Beside that, there is also evidence for directed manipulation influence of Twitter as “deviant” and inappropriate on Ira- of opinion with the help of social bots and fake accounts. nian domestic affairs2, most of the foreign news coverage So, it is obvious to ask, if there is an attempt to manipulate hailed Twitter to be “a new and influential medium for social the opinion-making process related to the Iranian protest in movements and international politics” (Burns and Eltham OSN by employing social bots, and how such manipulations 2009). Beside that, there was already a critical view upon will affect the discourse as a whole. Based on a sample of the influence of OSN’s as “tools” to express political opin- ca.
    [Show full text]
  • Political Astroturfing Across the World
    Political astroturfing across the world Franziska B. Keller∗ David Schochy Sebastian Stierz JungHwan Yang§ Paper prepared for the Harvard Disinformation Workshop Update 1 Introduction At the very latest since the Russian Internet Research Agency’s (IRA) intervention in the U.S. presidential election, scholars and the broader public have become wary of coordi- nated disinformation campaigns. These hidden activities aim to deteriorate the public’s trust in electoral institutions or the government’s legitimacy, and can exacerbate political polarization. But unfortunately, academic and public debates on the topic are haunted by conceptual ambiguities, and rely on few memorable examples, epitomized by the often cited “social bots” who are accused of having tried to influence public opinion in various contemporary political events. In this paper, we examine “political astroturfing,” a particular form of centrally co- ordinated disinformation campaigns in which participants pretend to be ordinary citizens acting independently (Kovic, Rauchfleisch, Sele, & Caspar, 2018). While the accounts as- sociated with a disinformation campaign may not necessarily spread incorrect information, they deceive the audience about the identity and motivation of the person manning the ac- count. And even though social bots have been in the focus of most academic research (Fer- rara, Varol, Davis, Menczer, & Flammini, 2016; Stella, Ferrara, & De Domenico, 2018), seemingly automated accounts make up only a small part – if any – of most astroturf- ing campaigns. For instigators of astroturfing, relying exclusively on social bots is not a promising strategy, as humans are good at detecting low-quality information (Pennycook & Rand, 2019). On the other hand, many bots detected by state-of-the-art social bot de- tection methods are not part of an astroturfing campaign, but unrelated non-political spam ∗Hong Kong University of Science and Technology yThe University of Manchester zGESIS – Leibniz Institute for the Social Sciences §University of Illinois at Urbana-Champaign 1 bots.
    [Show full text]
  • Understanding Users' Perspectives of News Bots in the Age of Social Media
    sustainability Article Utilizing Bots for Sustainable News Business: Understanding Users’ Perspectives of News Bots in the Age of Social Media Hyehyun Hong 1 and Hyun Jee Oh 2,* 1 Department of Advertising and Public Relations, Chung-Ang University, Seoul 06974, Korea; [email protected] 2 Department of Communication Studies, Hong Kong Baptist University, Kowloon Tong, Kowloon, Hong Kong SAR, China * Correspondence: [email protected] Received: 15 July 2020; Accepted: 10 August 2020; Published: 12 August 2020 Abstract: The move of news audiences to social media has presented a major challenge for news organizations. How to adapt and adjust to this social media environment is an important issue for sustainable news business. News bots are one of the key technologies offered in the current media environment and are widely applied in news production, dissemination, and interaction with audiences. While benefits and concerns coexist about the application of bots in news organizations, the current study aimed to examine how social media users perceive news bots, the factors that affect their acceptance of bots in news organizations, and how this is related to their evaluation of social media news in general. An analysis of the US national survey dataset showed that self-efficacy (confidence in identifying content from a bot) was a successful predictor of news bot acceptance, which in turn resulted in a positive evaluation of social media news in general. In addition, an individual’s perceived prevalence of social media news from bots had an indirect effect on acceptance by increasing self-efficacy. The results are discussed with the aim of providing a better understanding of news audiences in the social media environment, and practical implications for the sustainable news business are suggested.
    [Show full text]
  • Supplementary Materials For
    www.sciencemag.org/content/359/6380/1094/suppl/DC1 Supplementary Materials for The science of fake news David M. J. Lazer,*† Matthew A. Baum,* Yochai Benkler, Adam J. Berinsky, Kelly M. Greenhill, Filippo Menczer, Miriam J. Metzger, Brendan Nyhan, Gordon Pennycook, David Rothschild, Michael Schudson, Steven A. Sloman, Cass R. Sunstein, Emily A. Thorson, Duncan J. Watts, Jonathan L. Zittrain *These authors contributed equally to this work. †Corresponding author. Email: [email protected] Published 9 March 2018, Science 359, 1094 (2018) DOI: 10.1126/science.aao2998 This PDF file includes: List of Author Affiliations Supporting Materials References List of Author Affiliations David M. J. Lazer,1,2*† Matthew A. Baum,3* Yochai Benkler,4,5 Adam J. Berinsky,6 Kelly M. Greenhill,7,3 Filippo Menczer,8 Miriam J. Metzger,9 Brendan Nyhan,10 Gordon Pennycook,11 David Rothschild,12 Michael Schudson,13 Steven A. Sloman,14 Cass R. Sunstein,4 Emily A. Thorson,15 Duncan J. Watts,12 Jonathan L. Zittrain4,5 1Network Science Institute, Northeastern University, Boston, MA 02115, USA. 2Institute for Quantitative Social Science, Harvard University, Cambridge, MA 02138, USA 3John F. Kennedy School of Government, Harvard University, Cambridge, MA 02138, USA. 4Harvard Law School, Harvard University, Cambridge, MA 02138, USA. 5 Berkman Klein Center for Internet and Society, Cambridge, MA 02138, USA. 6Department of Political Science, Massachussets Institute of Technology, Cambridge, MA 02139, USA. 7Department of Political Science, Tufts University, Medford, MA 02155, USA. 8School of Informatics, Computing, and Engineering, Indiana University, Bloomington, IN 47405, USA. 9Department of Communication, University of California, Santa Barbara, Santa Barbara, CA 93106, USA.
    [Show full text]
  • Frame Viral Tweets.Pdf
    BIG IDEAS! Challenging Public Relations Research and Practice Viral Tweets, Fake News and Social Bots in Post-Factual Politics: The Case of the French Presidential Elections 2017 Alex Frame, Gilles Brachotte, Eric Leclercq, Marinette Savonnet [email protected] 20th Euprera Congress, Aarhus, 27-29th September 2018 Post-Factual Politics in the Age of Social Media 1. Algorithms based on popularity rather than veracity, linked to a business model based on views and likes, where novelty and sensationalism are of the essence; 2. Social trends of “whistle-blowing” linked to conspiracy theories and a pejorative image of corporate or institutional communication, which cast doubt on the neutrality of traditionally so-called expert figures, such as independent bodies or academics; 3. Algorithm-fuelled social dynamics on social networking sites which structure publics by similarity, leading to a fragmented digital public sphere where like-minded individuals congregate digitally, providing an “echo-chamber” effect susceptible to encourage even the most extreme political views and the “truths” upon which they are based. Fake News at the Presidential Elections Fake News at the Presidential Elections Fake News at the Presidential Elections Political Social Bots on Twitter Increasingly widespread around the world, over at least the last decade. Kremlin bot army (Lawrence Alexander on GlobalVoices.org; Stukal et al., 2017). DFRLab (Atlantic Council) Social Bots A “social bot” is: “a computer algorithm that automatically produces content and
    [Show full text]
  • Empirical Comparative Analysis of Bot Evidence in Social Networks
    Bots in Nets: Empirical Comparative Analysis of Bot Evidence in Social Networks Ross Schuchard1(&) , Andrew Crooks1,2 , Anthony Stefanidis2,3 , and Arie Croitoru2 1 Computational Social Science Program, Department of Computational and Data Sciences, George Mason University, Fairfax, VA 22030, USA [email protected] 2 Department of Geography and Geoinformation Science, George Mason University, Fairfax, VA 22030, USA 3 Criminal Investigations and Network Analysis Center, George Mason University, Fairfax, VA 22030, USA Abstract. The emergence of social bots within online social networks (OSNs) to diffuse information at scale has given rise to many efforts to detect them. While methodologies employed to detect the evolving sophistication of bots continue to improve, much work can be done to characterize the impact of bots on communication networks. In this study, we present a framework to describe the pervasiveness and relative importance of participants recognized as bots in various OSN conversations. Specifically, we harvested over 30 million tweets from three major global events in 2016 (the U.S. Presidential Election, the Ukrainian Conflict and Turkish Political Censorship) and compared the con- versational patterns of bots and humans within each event. We further examined the social network structure of each conversation to determine if bots exhibited any particular network influence, while also determining bot participation in key emergent network communities. The results showed that although participants recognized as social bots comprised only 0.28% of all OSN users in this study, they accounted for a significantly large portion of prominent centrality rankings across the three conversations. This includes the identification of individual bots as top-10 influencer nodes out of a total corpus consisting of more than 2.8 million nodes.
    [Show full text]
  • Disinformation, and Influence Campaigns on Twitter 'Fake News'
    Disinformation, ‘Fake News’ and Influence Campaigns on Twitter OCTOBER 2018 Matthew Hindman Vlad Barash George Washington University Graphika Contents Executive Summary . 3 Introduction . 7 A Problem Both Old and New . 9 Defining Fake News Outlets . 13 Bots, Trolls and ‘Cyborgs’ on Twitter . 16 Map Methodology . 19 Election Data and Maps . 22 Election Core Map Election Periphery Map Postelection Map Fake Accounts From Russia’s Most Prominent Troll Farm . 33 Disinformation Campaigns on Twitter: Chronotopes . 34 #NoDAPL #WikiLeaks #SpiritCooking #SyriaHoax #SethRich Conclusion . 43 Bibliography . 45 Notes . 55 2 EXECUTIVE SUMMARY This study is one of the largest analyses to date on how fake news spread on Twitter both during and after the 2016 election campaign. Using tools and mapping methods from Graphika, a social media intelligence firm, we study more than 10 million tweets from 700,000 Twitter accounts that linked to more than 600 fake and conspiracy news outlets. Crucially, we study fake and con- spiracy news both before and after the election, allowing us to measure how the fake news ecosystem has evolved since November 2016. Much fake news and disinformation is still being spread on Twitter. Consistent with other research, we find more than 6.6 million tweets linking to fake and conspiracy news publishers in the month before the 2016 election. Yet disinformation continues to be a substantial problem postelection, with 4.0 million tweets linking to fake and conspiracy news publishers found in a 30-day period from mid-March to mid-April 2017. Contrary to claims that fake news is a game of “whack-a-mole,” more than 80 percent of the disinformation accounts in our election maps are still active as this report goes to press.
    [Show full text]
  • Social Bots and Social Media Manipulation in 2020: the Year in Review
    Social Bots and Social Media Manipulation in 2020: The Year in Review Ho-Chun Herbert Chang∗a, Emily Chen∗b,c, Meiqing Zhang∗a, Goran Muric∗b, and Emilio Ferraraa,b,c aUSC Annenberg School of Communication bUSC Information Sciences Institute cUSC Department of Computer Science *These authors contributed equally to this work Abstract The year 2020 will be remembered for two events of global significance: the COVID-19 pandemic and 2020 U.S. Presidential Election. In this chapter, we summarize recent studies using large public Twitter data sets on these issues. We have three primary ob- jectives. First, we delineate epistemological and practical considerations when combin- ing the traditions of computational research and social science research. A sensible bal- ance should be struck when the stakes are high between advancing social theory and concrete, timely reporting of ongoing events. We additionally comment on the compu- tational challenges of gleaning insight from large amounts of social media data. Second, we characterize the role of social bots in social media manipulation around the discourse on the COVID-19 pandemic and 2020 U.S. Presidential Election. Third, we compare re- sults from 2020 to prior years to note that, although bot accounts still contribute to the emergence of echo-chambers, there is a transition from state-sponsored campaigns to do- mestically emergent sources of distortion. Furthermore, issues of public health can be confounded by political orientation, especially from localized communities of actors who spread misinformation. We conclude that automation and social media manipulation arXiv:2102.08436v1 [cs.SI] 16 Feb 2021 pose issues to a healthy and democratic discourse, precisely because they distort repre- sentation of pluralism within the public sphere.
    [Show full text]
  • The Rise of Social Botnets: Attacks and Countermeasures
    1 The Rise of Social Botnets: Attacks and Countermeasures Jinxue Zhang, Student Member, IEEE, Rui Zhang, Member, IEEE, Yanchao Zhang, Senior Member, IEEE, and Guanhua Yan Abstract—Online social networks (OSNs) are increasingly threatened by social bots which are software-controlled OSN accounts that mimic human users with malicious intentions. A social botnet refers to a group of social bots under the control of a single botmaster, which collaborate to conduct malicious behavior while mimicking the interactions among normal OSN users to reduce their individual risk of being detected. We demonstrate the effectiveness and advantages of exploiting a social botnet for spam distribution and digital-influence manipulation through real experiments on Twitter and also trace-driven simulations. We also propose the corresponding countermeasures and evaluate their effectiveness. Our results can help understand the potentially detrimental effects of social botnets and help OSNs improve their bot(net) detection systems. Index Terms—Social bot, social botnet, spam distribution, digital influence F 1 INTRODUCTION NLINE social networks (OSNs) are increasingly threat- two new social botnet attacks on Twitter, one of the most O ened by social bots [2] which are software-controlled OSN popular OSNs with over 302M monthly active users as of June accounts that mimic human users with malicious intentions. 2015 and over 500M new tweets daily. Then we propose two For example, according to a May 2012 article in Bloomberg defenses on the reported attacks, respectively. Our results help Businessweek,1 as many as 40% of the accounts on Facebook, understand the potentially detrimental effects of social botnets Twitter, and other popular OSNs are spammer accounts (or and shed the light for Twitter and other OSNs to improve their social bots), and about 8% of the messages sent via social bot(net) detection systems.
    [Show full text]
  • Supervised Machine Learning Bot Detection Techniques to Identify
    SMU Data Science Review Volume 1 | Number 2 Article 5 2018 Supervised Machine Learning Bot Detection Techniques to Identify Social Twitter Bots Phillip George Efthimion Southern Methodist University, [email protected] Scott aP yne Southern Methodist University, [email protected] Nicholas Proferes University of Kentucky, [email protected] Follow this and additional works at: https://scholar.smu.edu/datasciencereview Part of the Theory and Algorithms Commons Recommended Citation Efthimion, hiP llip George; Payne, Scott; and Proferes, Nicholas (2018) "Supervised Machine Learning Bot Detection Techniques to Identify Social Twitter Bots," SMU Data Science Review: Vol. 1 : No. 2 , Article 5. Available at: https://scholar.smu.edu/datasciencereview/vol1/iss2/5 This Article is brought to you for free and open access by SMU Scholar. It has been accepted for inclusion in SMU Data Science Review by an authorized administrator of SMU Scholar. For more information, please visit http://digitalrepository.smu.edu. Efthimion et al.: Supervised Machine Learning Bot Detection Techniques to Identify Social Twitter Bots Supervised Machine Learning Bot Detection Techniques to Identify Social Twitter Bots Phillip G. Efthimion1, Scott Payne1, Nick Proferes2 1Master of Science in Data Science, Southern Methodist University 6425 Boaz Lane, Dallas, TX 75205 {pefthimion, mspayne}@smu.edu [email protected] Abstract. In this paper, we present novel bot detection algorithms to identify Twitter bot accounts and to determine their prevalence in current online discourse. On social media, bots are ubiquitous. Bot accounts are problematic because they can manipulate information, spread misinformation, and promote unverified information, which can adversely affect public opinion on various topics, such as product sales and political campaigns.
    [Show full text]
  • You Must Be a Bot! How Beliefs Shape Twitter Profile Perceptions
    Disagree? You Must be a Bot! How Beliefs Shape Twitter Profile Perceptions MAGDALENA WISCHNEWSKI, University of Duisburg-Essen, Germany REBECCA BERNEMANN, University of Duisburg-Essen, Germany THAO NGO, University of Duisburg-Essen, Germany NICOLE KRÄMER, University of Duisburg-Essen, Germany In this paper, we investigate the human ability to distinguish political social bots from humans on Twitter. Following motivated reasoning theory from social and cognitive psychology, our central hypothesis is that especially those accounts which are opinion- incongruent are perceived as social bot accounts when the account is ambiguous about its nature. We also hypothesize that credibility ratings mediate this relationship. We asked N = 151 participants to evaluate 24 Twitter accounts and decide whether the accounts were humans or social bots. Findings support our motivated reasoning hypothesis for a sub-group of Twitter users (those who are more familiar with Twitter): Accounts that are opinion-incongruent are evaluated as relatively more bot-like than accounts that are opinion-congruent. Moreover, it does not matter whether the account is clearly social bot or human or ambiguous about its nature. This was mediated by perceived credibility in the sense that congruent profiles were evaluated to be more credible resulting in lower perceptions as bots. CCS Concepts: • Human-centered computing ! Empirical studies in HCI. Additional Key Words and Phrases: motivated reasoning, social bots, Twitter, credibility, partisanship, bias ACM Reference Format: Magdalena Wischnewski, Rebecca Bernemann, Thao Ngo, and Nicole Krämer. 2021. Disagree? You Must be a Bot! How Beliefs Shape Twitter Profile Perceptions. In CHI Conference on Human Factors in Computing Systems (CHI ’21), May 8–13, 2021, Yokohama, Japan.
    [Show full text]
  • Introduction: What Is Computational Propaganda?
    !1 Introduction: What Is Computational Propaganda? Digital technologies hold great promise for democracy. Social media tools and the wider resources of the Internet o"er tremendous access to data, knowledge, social networks, and collective engagement opportunities, and can help us to build better democracies (Howard, 2015; Margetts et al., 2015). Unwelcome obstacles are, however, disrupting the creative democratic applications of information technologies (Woolley, 2016; Gallacher et al., 2017; Vosoughi, Roy, & Aral, 2018). Massive social platforms like Facebook and Twitter are struggling to come to grips with the ways their creations can be used for political control. Social media algorithms may be creating echo chambers in which public conversations get polluted and polarized. Surveillance capabilities are outstripping civil protections. Political “bots” (software agents used to generate simple messages and “conversations” on social media) are masquerading as genuine grassroots movements to manipulate public opinion. Online hate speech is gaining currency. Malicious actors and digital marketers run junk news factories that disseminate misinformation to harm opponents or earn click-through advertising revenue.# It is no exaggeration to say that coordinated e"orts are even now working to seed chaos in many political systems worldwide. Some militaries and intelligence agencies are making use of social media as conduits to undermine democratic processes and bring down democratic institutions altogether (Bradshaw & Howard, 2017). Most democratic governments are preparing their legal and regulatory responses. But unintended consequences from over- regulation, or regulation uninformed by systematic research, may be as damaging to democratic systems as the threats themselves.# We live in a time of extraordinary political upheaval and change, with political movements and parties rising and declining rapidly (Kreiss, 2016; Anstead, 2017).
    [Show full text]