arXiv subjects

Kathleen M. Carley

Publications and source records attributed to Kathleen M. Carley.

At least 73 records · Page 4Linked to original sources

arXiv2021

Analysis of External Content in the Vaccination Discussion on Twitter

The spread of coronavirus and anti-vaccine conspiracies online hindered public health responses to the pandemic. We examined the content of external articles shared on Twitter from February to June 2020 to understand how conspiracy theories and fake news competed with legitimate sources of information. Examining external content--articles, rather than social media posts--is a novel methodology that allows for non-social media specific analysis of misinformation, tracking of changing narratives over time, and determining which types of resources (government, news, scientific, or dubious) dominate the pandemic vaccine conversation. We find that distinct narratives emerge, those narratives change over time, and lack of government and scientific messaging on coronavirus created an information vacuum filled by both traditional news and conspiracy theories.

cs.SI↗

arXiv2021

Coordinating Narratives and the Capitol Riots on Parler

Coordinated disinformation campaigns are used to influence social media users, potentially leading to offline violence. In this study, we introduce a general methodology to uncover coordinated messaging through analysis of user parleys on Parler. The proposed method constructs a user-to-user coordination network graph induced by a user-to-text graph and a text-to-text similarity graph. The text-to-text graph is constructed based on the textual similarity of Parler posts. We study three influential groups of users in the 6 January 2020 Capitol riots and detect networks of coordinated user clusters that are all posting similar textual content in support of different disinformation narratives related to the U.S. 2020 elections.

cs.SI↗

arXiv2021

Climate Change Conspiracy Theories on Social Media

One of the critical emerging challenges in climate change communication is the prevalence of conspiracy theories. This paper discusses some of the major conspiracy theories related to climate change found in a large Twitter corpus. We use a state-of-the-art stance detection method to find whether conspiracy theories are more popular among Disbelievers or Believers of climate change. We then analyze which conspiracy theory is more popular than the others and how popularity changes with climate change belief. We find that Disbelievers of climate change are overwhelmingly responsible for sharing messages with conspiracy theory-related keywords, and not all conspiracy theories are equally shared. Lastly, we discuss the implications of our findings for climate change communication.

cs.SI↗

arXiv2021

A Synchronized Action Framework for Responsible Detection of Coordination on Social Media

The study of coordinated manipulation of conversations on social media has become more prevalent as social media's role in amplifying misinformation, hate, and polarization has come under scrutiny. We discuss the implications of successful coordination detection algorithms based on shifts of power, and consider how responsible coordination detection may be carried out through synchronized action. We then propose a Synchronized Action Framework for detection of automated coordination through construction and analysis of multi-view networks. We validate our framework by examining the Reopen America conversation on Twitter, discovering three coordinated campaigns. We further investigate covert coordination surrounding the protests and find the task to be far more complex than examples seen in prior work, demonstrating the need for our multi-view approach. A cluster of suspicious users is identified and the activity of three members is detailed. These users amplify protest messages using the same hashtags at very similar times, though they all focus on different states. Through this analysis, we emphasize both the potential usefulness of coordination detection algorithms in investigating amplification, and the need for careful and responsible deployment of such tools.

cs.SI↗

arXiv2021

Learning future terrorist targets through temporal meta-graphs

In the last 20 years, terrorism has led to hundreds of thousands of deaths and massive economic, political, and humanitarian crises in several regions of the world. Using real-world data on attacks occurred in Afghanistan and Iraq from 2001 to 2018, we propose the use of temporal meta-graphs and deep learning to forecast future terrorist targets. Focusing on three event dimensions, i.e., employed weapons, deployed tactics and chosen targets, meta-graphs map the connections among temporally close attacks, capturing their operational similarities and dependencies. From these temporal meta-graphs, we derive 2-day-based time series that measure the centrality of each feature within each dimension over time. Formulating the problem in the context of the strategic behavior of terrorist actors, these multivariate temporal sequences are then utilized to learn what target types are at the highest risk of being chosen. The paper makes two contributions. First, it demonstrates that engineering the feature space via temporal meta-graphs produces richer knowledge than shallow time-series that only rely on frequency of feature occurrences. Second, the performed experiments reveal that bi-directional LSTM networks achieve superior forecasting performance compared to other algorithms, calling for future research aiming at fully discovering the potential of artificial intelligence to counter terrorist violence.

cs.LG↗

arXiv2021

Exploring Polarization of Users Behavior on Twitter During the 2019 South American Protests

Research across different disciplines has documented the expanding polarization in social media. However, much of it focused on the US political system or its culturally controversial topics. In this work, we explore polarization on Twitter in a different context, namely the protest that paralyzed several countries in the South American region in 2019. By leveraging users' endorsement of politicians' tweets and hashtag campaigns with defined stances towards the protest (for or against), we construct a weakly labeled stance dataset with millions of users. We explore polarization in two related dimensions: language and news consumption patterns. In terms of linguistic polarization, we apply recent insights that leveraged machine translation methods, showing that the two communities speak consistently "different" languages, mainly along ideological lines (e.g., fascist translates to communist). Our results indicate that this recently-proposed methodology is also informative in different languages and contexts than originally applied. In terms of news consumption patterns, we cluster news agencies based on homogeneity of their user bases and quantify the observed polarization in its consumption. We find empirical evidence of the "filter bubble" phenomenon during the event, as we not only show that the user bases are homogeneous in terms of stance, but the probability that a user transitions from media of different clusters is low.

cs.SI↗

arXiv2021

The Coronavirus is a Bioweapon: Analysing Coronavirus Fact-Checked Stories

The 2020 coronavirus pandemic has heightened the need to flag coronavirus-related misinformation, and fact-checking groups have taken to verifying misinformation on the Internet. We explore stories reported by fact-checking groups PolitiFact, Poynter and Snopes from January to June 2020, characterising them into six story clusters before then analyse time-series and story validity trends and the level of agreement across sites. We further break down the story clusters into more granular story types by proposing a unique automated method with a BERT classifier, which can be used to classify diverse story sources, in both fact-checked stories and tweets.

cs.SI↗

arXiv2021

A Weakly Supervised Approach for Classifying Stance in Twitter Replies

Conversations on social media (SM) are increasingly being used to investigate social issues on the web, such as online harassment and rumor spread. For such issues, a common thread of research uses adversarial reactions, e.g., replies pointing out factual inaccuracies in rumors. Though adversarial reactions are prevalent in online conversations, inferring those adverse views (or stance) from the text in replies is difficult and requires complex natural language processing (NLP) models. Moreover, conventional NLP models for stance mining need labeled data for supervised learning. Getting labeled conversations can itself be challenging as conversations can be on any topic, and topics change over time. These challenges make learning the stance a difficult NLP problem. In this research, we first create a new stance dataset comprised of three different topics by labeling both users' opinions on the topics (as in pro/con) and users' stance while replying to others' posts (as in favor/oppose). As we find limitations with supervised approaches, we propose a weakly-supervised approach to predict the stance in Twitter replies. Our novel method allows using a smaller number of hashtags to generate weak labels for Twitter replies. Compared to supervised learning, our method improves the mean F1-macro by 8\% on the hand-labeled dataset without using any hand-labeled examples in the training set. We further show the applicability of our proposed method on COVID 19 related conversations on Twitter.

cs.CL↗

arXiv2021

Measuring Node Contribution to Community Structure with Modularity Vitality

Community-aware centrality is an emerging research area in network science concerned with the importance of nodes in relation to community structure. Measures are a function of a network's structure and a given partition. Previous approaches extend classical centrality measures to account for community structure with little connection to community detection theory. In contrast, we propose cluster-quality vitality measures, i.e., modularity vitality, a community-aware measure which is well-grounded in both centrality and community detection theory. Modularity vitality quantifies positive and negative contributions to community structure, which indicate a node's role as a community bridge or hub. We derive a computationally efficient method of calculating modularity vitality for all nodes in O(M + NC) time, where C is the number of communities. We systematically fragment networks by removing central nodes, and find that modularity vitality consistently outperforms existing community-aware centrality measures. Modularity vitality is over 8 times more effective than the next-best method on a million-node infrastructure network. This result does not generalize to social media communication networks, which exhibit extreme robustness to all community-aware centrality attacks. This robustness suggests that user-based interventions to mitigate misinformation diffusion will be ineffective. Finally, we demonstrate that modularity vitality provides a new approach to community-deception.

cs.SI↗

arXiv2020

Characterizing Sociolinguistic Variation in the Competing Vaccination Communities

Public health practitioners and policy makers grapple with the challenge of devising effective message-based interventions for debunking public health misinformation in cyber communities. "Framing" and "personalization" of the message is one of the key features for devising a persuasive messaging strategy. For an effective health communication, it is imperative to focus on "preference-based framing" where the preferences of the target sub-community are taken into consideration. To achieve that, it is important to understand and hence characterize the target sub-communities in terms of their social interactions. In the context of health-related misinformation, vaccination remains to be the most prevalent topic of discord. Hence, in this paper, we conduct a sociolinguistic analysis of the two competing vaccination communities on Twitter: "pro-vaxxers" or individuals who believe in the effectiveness of vaccinations, and "anti-vaxxers" or individuals who are opposed to vaccinations. Our data analysis show significant linguistic variation between the two communities in terms of their usage of linguistic intensifiers, pronouns, and uncertainty words. Our network-level analysis show significant differences between the two communities in terms of their network density, echo-chamberness, and the EI index. We hypothesize that these sociolinguistic differences can be used as proxies to characterize and understand these communities to devise better message interventions.

cs.SI↗

arXiv2020

Characterizing COVID-19 Misinformation Communities Using a Novel Twitter Dataset

From conspiracy theories to fake cures and fake treatments, COVID-19 has become a hot-bed for the spread of misinformation online. It is more important than ever to identify methods to debunk and correct false information online. In this paper, we present a methodology and analyses to characterize the two competing COVID-19 misinformation communities online: (i) misinformed users or users who are actively posting misinformation, and (ii) informed users or users who are actively spreading true information, or calling out misinformation. The goals of this study are two-fold: (i) collecting a diverse set of annotated COVID-19 Twitter dataset that can be used by the research community to conduct meaningful analysis; and (ii) characterizing the two target communities in terms of their network structure, linguistic patterns, and their membership in other communities. Our analyses show that COVID-19 misinformed communities are denser, and more organized than informed communities, with a possibility of a high volume of the misinformation being part of disinformation campaigns. Our analyses also suggest that a large majority of misinformed users may be anti-vaxxers. Finally, our sociolinguistic analyses suggest that COVID-19 informed users tend to use more narratives than misinformed users.

cs.SI↗

arXiv2020

Affective Polarization in Online Climate Change Discourse on Twitter

Online social media has become an important platform to organize around different socio-cultural and political topics. An extensive scholarship has discussed how people are divided into echo-chamber-like groups. However, there is a lack of work related to quantifying hostile communication or \textit{affective polarization} between two competing groups. This paper proposes a systematic, network-based methodology for examining affective polarization in online conversations. Further, we apply our framework to 100 weeks of Twitter discourse about climate change. We find that deniers of climate change (Disbelievers) are more hostile towards people who believe (Believers) in the anthropogenic cause of climate change than vice versa. Moreover, Disbelievers use more words and hashtags related to natural disasters during more hostile weeks as compared to Believers. These findings bear implications for studying affective polarization in online discourse, especially concerning the subject of climate change. Lastly, we discuss our findings in the context of increasingly important climate change communication research.

cs.SI↗

arXiv2020

Polarizing Tweets on Climate Change

We introduce a framework to analyze the conversation between two competing groups of Twitter users, one who believe in the anthropogenic causes of climate change (Believers) and a second who are skeptical (Disbelievers). As a case study, we use Climate Change related tweets during the United Nation's (UN) Climate Change Conference - COP24 (2018), Katowice, Poland. We find that both Disbelievers and Believers talk within their group more than with the other group; this is more so the case for Disbelievers than for Believers. The Disbeliever messages focused more on attacking those personalities that believe in the anthropogenic causes of climate change. On the other hand, Believer messages focused on calls to combat climate change. We find that in both Disbelievers and Believers bot-like accounts were equally active and that unlike Believers, Disbelievers get their news from a concentrated number of news sources.

cs.SI↗

arXiv2020

Social Cybersecurity Chapter 13: Casestudy with COVID-19 Pandemic

The purpose of this case study is to leverage the concepts and tools presented in the preceding chapters and apply them in a real world social cybersecurity context. With the COVID-19 pandemic emerging as a defining event of the 21st Century and a magnet for disinformation maneuver, we have selected the pandemic and its related social media conversation to focus our efforts on. This chapter therefore applies the tools of information operation maneuver, bot detection and characterization, meme detection and characterization, and information mapping to the COVID-19 related conversation on Twitter. This chapter uses these tools to analyze a stream containing 206 million tweets from 27 million unique users from 15 March 2020 to 30 April 2020. Our results shed light on elaborate information operations that leverage the full breadth of the BEND maneuvers and use bots for important shaping operations.

cs.CY↗

arXiv2020

Characterizing Communities of Hashtag Usage on Twitter During the 2020 COVID-19 Pandemic by Multi-view Clustering

The COVID-19 pandemic has produced a flurry of online activity on social media sites. As such, analysis of social media data during the COVID-19 pandemic can produce unique insights into discussion topics and how those topics evolve over the course of the pandemic. In this study, we propose analyzing discussion topics on Twitter by clustering hashtags. In order to obtain high-quality clusters of the Twitter hashtags, we also propose a novel multi-view clustering technique that incorporates multiple different data types that can be used to describe how users interact with hashtags. The results of our multi-view clustering show that there are distinct temporal and topical trends present within COVID-19 twitter discussion. In particular, we find that some topical clusters of hashtags shift over the course of the pandemic, while others are persistent throughout, and that there are distinct temporal trends in hashtag usage. This study is the first to use multi-view clustering to analyze hashtags and the first analysis of the greater trends of discussion occurring online during the COVID-19 pandemic.

cs.SI↗

arXiv2020

A Computational Analysis of Polarization on Indian and Pakistani Social Media

Between February 14, 2019 and March 4, 2019, a terrorist attack in Pulwama, Kashmir followed by retaliatory airstrikes led to rising tensions between India and Pakistan, two nuclear-armed countries. In this work, we examine polarizing messaging on Twitter during these events, particularly focusing on the positions of Indian and Pakistani politicians. We use a label propagation technique focused on hashtag co-occurrences to find polarizing tweets and users. Our analysis reveals that politicians in the ruling political party in India (BJP) used polarized hashtags and called for escalation of conflict more so than politicians from other parties. Our work offers the first analysis of how escalating tensions between India and Pakistan manifest on Twitter and provides a framework for studying polarizing messages.

cs.SI↗

arXiv2020

Bot-Match: Social Bot Detection with Recursive Nearest Neighbors Search

Social bots have emerged over the last decade, initially creating a nuisance while more recently used to intimidate journalists, sway electoral events, and aggravate existing social fissures. This social threat has spawned a bot detection algorithms race in which detection algorithms evolve in an attempt to keep up with increasingly sophisticated bot accounts. This cat and mouse cycle has illuminated the limitations of supervised machine learning algorithms, where researchers attempt to use yesterday's data to predict tomorrow's bots. This gap means that researchers, journalists, and analysts daily identify malicious bot accounts that are undetected by state of the art supervised bot detection algorithms. These analysts often desire to find similar bot accounts without labeling/training a new model, where similarity can be defined by content, network position, or both. A similarity based algorithm could complement existing supervised and unsupervised methods and fill this gap. To this end, we present the Bot-Match methodology in which we evaluate social media embeddings that enable a semi-supervised recursive nearest neighbors search to map an emerging social cybersecurity threat given one or more seed accounts.

cs.SI↗

arXiv2020

Stance in Replies and Quotes (SRQ): A New Dataset For Learning Stance in Twitter Conversations

Automated ways to extract stance (denying vs. supporting opinions) from conversations on social media are essential to advance opinion mining research. Recently, there is a renewed excitement in the field as we see new models attempting to improve the state-of-the-art. However, for training and evaluating the models, the datasets used are often small. Additionally, these small datasets have uneven class distributions, i.e., only a tiny fraction of the examples in the dataset have favoring or denying stances, and most other examples have no clear stance. Moreover, the existing datasets do not distinguish between the different types of conversations on social media (e.g., replying vs. quoting on Twitter). Because of this, models trained on one event do not generalize to other events. In the presented work, we create a new dataset by labeling stance in responses to posts on Twitter (both replies and quotes) on controversial issues. To the best of our knowledge, this is currently the largest human-labeled stance dataset for Twitter conversations with over 5200 stance labels. More importantly, we designed a tweet collection methodology that favors the selection of denial-type responses. This class is expected to be more useful in the identification of rumors and determining antagonistic relationships between users. Moreover, we include many baseline models for learning the stance in conversations and compare the performance of various models. We show that combining data from replies and quotes decreases the accuracy of models indicating that the two modalities behave differently when it comes to stance learning.

cs.CL↗