The role of social media in promoting media pluralism was initially viewed as wholly positive. However, some governments are allegedly manipulating social media by hiring online commentators (also known as trolls) to spread propaganda and disinformation. In particular, an alleged system of professional trolls operating both domestically and internationally exists in Russia. In 2018, Twitter released data on accounts identified as Russian trolls, starting a wave of research. However, while foreign-targeted English language operations of these trolls have received significant attention, no research has analyzed their Russian language domestic and regional-targeted activities. We address this gap by characterizing the Russian-language operations of Russian trolls. We first perform a descriptive analysis, and then focus in on the trolls' operation related to the crash of Malaysia Airlines flight MH17.
Among other things, we find that Russian-language trolls have run 163 hashtag campaigns (where hashtag use grows abruptly within a month). The main political sentiments of such campaigns were praising Russia and Putin (29%), criticizing Ukraine (26%), and criticizing the United States and Obama (9%). Further, trolls actively reshared information with 76% of tweets being retweets or containing a URL. Additionally, we observe periodic temporal patterns of tweeting suggesting that trolls use automation tools. Further, we find that trolls' information campaign on the MH17 crash was the largest in terms of tweet count. However, around 68% of tweets posted with MH17 hashtags were likely used simply for hashtag amplification. With these tweets excluded, about 49% of the tweets suggested to varying levels that Ukraine was responsible for the crash, and only 13% contained disinformation and propaganda presented as news. Interestingly, trolls promoted inconsistent alternative theories for the crash.
There is evidence that Russia's Internet Research Agency attempted to interfere with the 2016 U.S. election by running fake accounts on Twitter - often referred to as "Russian trolls". In this work, we: 1) develop machine learning models that predict whether a Twitter account is a Russian troll within a set of 170K control accounts; and, 2) demonstrate that it is possible to use this model to find active accounts on Twitter still likely acting on behalf of the Russian state. Using both behavioral and linguistic features, we show that it is possible to distinguish between a troll and a non-trol…
In the digital era, individuals are increasingly profiled and grouped based on the traces they leave behind in online social networks such as Twitter and Facebook. In this paper, we develop and evaluate a novel text analysis approach for studying user identity and social roles by redefining identity as a sequence of timestamped items (e.g. tweet texts). We operationalise this idea by developing a novel text distance metric, the time-sensitive semantic edit distance (t-SED), which accounts for the temporal context across multiple traces. To evaluate this method we undertake a case study of Russ…
State-sponsored organizations are increasingly linked to efforts aimed to exploit social media for information warfare and manipulating public opinion. Typically, their activities rely on a number of social network accounts they control, aka trolls, that post and interact with other users disguised as "regular" users. These accounts often use images and memes, along with textual content, in order to increase the engagement and the credibility of their posts. In this paper, we present the first study of images shared by state-sponsored accounts by analyzing a ground truth dataset of 1.8M images…
Until recently, social media was seen to promote democratic discourse on social and political issues. However, this powerful communication platform has come under scrutiny for allowing hostile actors to exploit online discussions in an attempt to manipulate public opinion. A case in point is the ongoing U.S. Congress' investigation of Russian interference in the 2016 U.S. election campaign, with Russia accused of using trolls (malicious accounts created to manipulate) and bots to spread misinformation and politically biased information. In this study, we explore the effects of this manipulatio…
University of Southern California（南加州大学）；Information Sciences Institute（信息科学研究所）；University of Applied Sciences and Arts of Southern Switzerland (SUPSI)（瑞士南部应用科学与艺术大学（SUPSI））；University of Bern（伯尔尼大学）
Since the 2016 US Presidential election, social media abuse has been eliciting massive concern in the academic community and beyond. Preventing and limiting the malicious activity of users, such as trolls and bots, in their manipulation campaigns is of paramount importance for the integrity of democracy, public health, and more. However, the automated detection of troll accounts is an open challenge. In this work, we propose an approach based on Inverse Reinforcement Learning (IRL) to capture troll behavior and identify troll accounts. We employ IRL to infer a set of online incentives that may…
In this analysis, we work with the data set that was compiled by Darren Linvill and Patrick Warren, along with a representative sample of Facebook ads that were released by the House Intelligence Committee Minority. The goal of this analysis is to use the categories defined by Linvill and Warren in the Twitter data and investigate if these categories exist in Facebook ads. This begin to give us insights to the tactics used between the two social media services. Further, we try to replicate Linvill and Warren's original categorization of the Twitter data. Lastly, we investigate what categories …
We developed and used a collection of statistical methods (unsupervised machine learning) to extract relevant information from a Twitter supplied data set consisting of alleged Russian trolls who (allegedly) attempted to influence the 2016 US Presidential election. These unsupervised statistical methods allow fast identification of (i) emergent language communities within the troll population, (ii) the transnational scope of the operation and (iii) operational characteristics of trolls that can be used for future identification. Using natural language processing, manifold learning and Fourier …
Until recently, social media were seen to promote democratic discourse on social and political issues. However, this powerful communication ecosystem has come under scrutiny for allowing hostile actors to exploit online discussions in an attempt to manipulate public opinion. A case in point is the ongoing U.S. Congress investigation of Russian interference in the 2016 U.S. election campaign, with Russia accused of, among other things, using trolls (malicious accounts created for the purpose of manipulation) and bots (automated accounts) to spread propaganda and politically biased information. …
Universitat Politècnica de València（瓦伦西亚理工大学）；CNRS（法国国家科学研究中心）；Université Paris 13（巴黎第十三大学）
The online new emerging suspicious users, that usually are called trolls, are one of the main sources of hate, fake, and deceptive online messages. Some agendas are utilizing these harmful users to spread incitement tweets, and as a consequence, the audience get deceived. The challenge in detecting such accounts is that they conceal their identities which make them disguised in social media, adding more difficulty to identify them using just their social network information. Therefore, in this paper, we propose a text-based approach to detect the online trolls such as those that were discovere…
In a review into automated and malicious activity Twitter released a list of accounts that they believed were connected to state sponsored manipulation of the 2016 American Election. This list details 2,752 accounts Twitter believed to be controlled by Russian operatives. In the absence of a similar list of operatives active within the debate on the 2016 UK referendum on membership of the European Union (Brexit) we investigated the behaviour of the same American Election focused accounts in the production of content related to the UK-EU referendum. We found that within our dataset we had Brexi…
Friedrich Schiller University Jena（弗里德里希·席勒大学耶拿分校）
The current study yielded a number of important findings. We managed to build a neural network that achieved an accuracy score of 91 per cent in classifying troll and genuine tweets. By means of regression analysis, we identified a number of features that make a tweet more susceptible to correct labelling and found that they are inherently present in troll tweets as a special type of discourse. We hypothesised that those features are grounded in the sociolinguistic limitations of troll writing, which can be best described as a combination of two factors: speaking with a purpose and trying to m…
We investigate the political roles of "Internet trolls" in social media. Political trolls, such as the ones linked to the Russian Internet Research Agency (IRA), have recently gained enormous attention for their ability to sway public opinion and even influence elections. Analysis of the online traces of trolls has shown different behavioral patterns, which target different slices of the population. However, this analysis is manual and labor-intensive, thus making it impractical as a first-response tool for newly-discovered troll farms. In this paper, we show how to automate this analysis by u…
Over the past couple of years, anecdotal evidence has emerged linking coordinated campaigns by state-sponsored actors with efforts to manipulate public opinion on the Web, often around major political events, through dedicated accounts, or "trolls." Although they are often involved in spreading disinformation on social media, there is little understanding of how these trolls operate, what type of content they disseminate, and most importantly their influence on the information ecosystem. In this paper, we shed light on these questions by analyzing 27K tweets posted by 1K Twitter users identifi…
Being a matter of cognition, user interests should be apt to classification independent of the language of users, social network and content of interest itself. To prove it, we analyze a collection of English and Russian Twitter and Vkontakte community pages by interests of their followers. First, we create a model of Major Interests (MaIs) with the help of expert analysis and then classify a set of pages using machine learning algorithms (SVM, Neural Network, Naive Bayes, and some other). We take three interest domains that are typical of both English and Russian-speaking communities: footbal…
Cyberbullying or Online harassment detection on social media for various major languages is currently being given a good amount of focus by researchers worldwide. Being the seventh most speaking language in the world and increasing usage of online platform among the Bengali speaking people urge to find effective detection technique to handle the online harassment. In this paper, we have proposed binary and multiclass classification model using hybrid neural network for bully expression detection in Bengali language. We have used 44,001 users comments from popular public Facebook pages, which f…
The (((echo))) symbol -- triple parenthesis surrounding a name, made it to mainstream social networks in early 2016, with the intensification of the U.S. Presidential race. It was used by members of the alt-right, white supremacists and internet trolls to tag people of Jewish heritage -- a modern incarnation of the infamous yellow badge (Judenstern) used in Nazi-Germany. Tracking this trending meme, its meaning, and its function has proved elusive for its semantic ambiguity (e.g., a symbol for a virtual hug). In this paper we report of the construction of an appropriate dataset allowing the re…
