Metaphor, discussing one concept in terms of another, is abundant in politics and can shape how people understand important issues. We develop a computational approach to measure metaphorical language, focusing on immigration discourse on social media. Grounded in qualitative social science research, we identify seven concepts evoked in immigration discourse (e.g. water or vermin). We propose and evaluate a novel technique that leverages both word-level and document-level signals to measure metaphor with respect to these concepts. We then study the relationship between metaphor, political ideology, and user engagement in 400K US tweets about immigration. While conservatives tend to use dehumanizing metaphors more than liberals, this effect varies widely across concepts. Moreover, creature-related metaphor is associated with more retweets, especially for liberal authors. Our work highlights the potential for computational methods to complement qualitative approaches in understanding subtle and implicit language in political discourse.

Peer Reviewed

We introduce the Deep Value Benchmark (DVB), an evaluation framework that directly tests whether large language models (LLMs) learn fundamental human values or merely surface-level preferences. This distinction is critical for AI alignment: Systems that capture deeper values are likely to generalize human intentions robustly, while those that capture only superficial patterns in preference data risk producing misaligned behavior. The DVB uses a novel experimental design with controlled confounding between deep values (e.g., moral principles) and shallow features (e.g., superficial attributes). In the training phase, we expose LLMs to human preference data with deliberately correlated deep and shallow features---for instance, where a user consistently prefers (non-maleficence, formal language) options over (justice, informal language) alternatives. The testing phase then breaks these correlations, presenting choices between (justice, formal language) and (non-maleficence, informal language) options. This design allows us to precisely measure a model's Deep Value Generalization Rate (DVGR)---the probability of generalizing based on the underlying value rather than the shallow feature. Across 9 different models, the average DVGR is just 0.30. All models generalize deep values less than chance. Larger models have a (slightly) lower DVGR than smaller models. We are releasing our dataset, which was subject to three separate human validation experiments. DVB provides an interpretable measure of a core feature of alignment.

How do Wikipedians maintain an accurate encyclopedia during an ongoing geopolitical conflict where state actors might seek to spread disinformation or conduct an information operation? In the context of the Russia-Ukraine War, this question becomes more pressing, given the Russian government's extensive history of orchestrating information campaigns. We conducted an interview study with 13 expert Wikipedians involved in the Russo-Ukrainian War topic area on the English-language edition of Wikipedia. While our participants did not perceive there to be clear evidence of a state-backed information operation, they agreed that war-related articles experienced high levels of disruptive editing from both Russia-aligned and Ukraine-aligned accounts. The English-language edition of Wikipedia had existing policies and processes at its disposal to counter such disruption. State-backed or not, the disruptive activity created time-intensive maintenance work for our participants. Finally, participants considered English-language Wikipedia to be more resilient than social media in preventing the spread of false information online. We conclude by discussing sociotechnical implications for Wikipedia and social platforms.

Intermedia agenda setting (IAS) theory suggests that different news sources can influence each other's agenda. While this theory has been well-established in existing literature, whether it still holds in today's high-choice media environment, which includes news producers of different credibility and ideology dispositions, is an open question. Through two case studies--the 2016 and 2020 U.S. presidential elections--we show that media are still largely aligned, especially in broad topics they choose to cover, and that the level of alignment along the credibility dimension is comparable to that along the ideology dimension. Furthermore, we find that the coverage of the Republican candidate is better aligned across different media types than that of the Democratic candidate, and that media divergence has increased along both dimensions from 2016 to 2020. Finally, we demonstrate that high-credibility media still plays a dominant role in the IAS process, yet with a cautious warning of its declining IAS power for the Democratic candidate over the course of four years.

Many studies explore how people "come into" misinformation exposure. But much less is known about how people "come out of" misinformation exposure. Do people organically sever ties to misinformation spreaders? And what predicts doing so? Over six months, we tracked the frequency and predictors of ~900K followers unfollowing ~5K health misinformation spreaders on Twitter. We found that misinformation ties are persistent. Monthly unfollowing rates are just 0.52%. In other words, 99.5% of misinformation ties persist each month. Users are also 31% more likely to unfollownon- misinformation spreaders than they are to unfollow misinformation spreaders. Although generally infrequent, the factors most associated with unfollowing misinformation spreaders are (1) redundancy and (2) ideology. First, users initially following many spreaders, or who follow spreaders that tweet often, are most likely to unfollow later. Second, liberals are more likely to unfollow than conservatives. Overall, we observe a strong persistence of misinformation ties. The fact that users rarely unfollow misinformation spreaders suggests a need for external nudges and the importance of preventing exposure from arising in the first place.

Social media enables the rapid spread of many kinds of information, from pop culture memes to social movements. However, little is known about how information crosses linguistic boundaries. We apply causal inference techniques on the European Twitter network to quantify the structural role and communication influence of multilingual users in cross-lingual information exchange. Overall, multilinguals play an essential role; posting in multiple languages increases betweenness centrality by 13%, and having a multilingual network neighbor increases monolinguals’ odds of sharing domains and hashtags from another language 16-fold and 4-fold, respectively. We further show that multilinguals have a greater impact on diffusing information is less accessible to their monolingual compatriots, such as information from far-away countries and content about regional politics, nascent social movements, and job opportunities. By highlighting information exchange across borders, this work sheds light on a crucial component of how information and ideas spread around the world.

Past work has explored various ways for online platforms to leverage crowd wisdom for misinformation detection and moderation. Yet, platforms often relegate governance to their communities, and limited research has been done from the perspective of these communities and their moderators. How is misinformation currently moderated in online communities that are heavily self-governed? What role does the crowd play in this process, and how can this process be improved? In this study, we answer these questions through semi-structured interviews with Reddit moderators. We focus on a case study of COVID-19 misinformation. First, our analysis identifies a general moderation workflow model encompassing various processes participants use for handling COVID-19 misinformation. Further, we show that the moderation workflow revolves around three elements: content facticity, user intent, and perceived harm. Next, our interviews reveal that Reddit moderators rely on two types of crowd wisdom for misinformation detection. Almost all participants are heavily reliant on reports from crowds of ordinary users to identify potential misinformation. A second crowd--participants' own moderation teams and expert moderators of other communities--provide support when participants encounter difficult, ambiguous cases. Finally, we use design probes to better understand how different types of crowd signals---from ordinary users and moderators---readily available on Reddit can assist moderators with identifying misinformation. We observe that nearly half of all participants preferred these cues over labels from expert fact-checkers because these cues can help them discern user intent. Additionally, a quarter of the participants distrust professional fact-checkers, raising important concerns about misinformation moderation.

The past decade in the US has been one of the most politically polarizing in recent memory. Ordinary Democrats and Republicans fundamentally dislike and distrust each other, even when they agree on policy issues. This increase in hostility towards opposing party supporters, commonly called affective polarization, has important ramifications that threaten democracy. Political science research suggests that at least part of this polarization stems from Democrats' misperceptions about Republicans' political views and vice-versa. Therefore, in this work, drawing on insights from political science and game studies research, we designed an online casual game that combines the relaxed, playful nonpartisan norms of casual games with corrective information about party supporters' political views that are often misperceived. Through an experiment, we found that playing the game significantly reduces negative feelings toward outparty supporters among Democrats, but not Republicans. It was also effective in improving willingness to talk politics with outparty supporters. Further, we identified psychological reactance as a potential mechanism that affects the effectiveness of depolarization interventions. Finally, our analyses suggest that the game versions with political content were rated to be just as fun to play as a game version without any political content suggesting that, contrary to popular belief, people do like to mix politics and play.

The framing of political issues can influence policy and public opinion.Even though the public plays a key role in creating and spreading frames, little is known about how ordinary people on social media frame political issues.By creating a new dataset of immigrationrelated tweets labeled for multiple framing typologies from political communication theory, we develop supervised models to detect frames.We demonstrate how users' ideology and region impact framing choices, and how a message's framing influences audience responses.We find that the more commonlyused issue-generic frames obscure important ideological and regional patterns that are only revealed by immigration-specific frames.Furthermore, frames oriented towards human interests, culture, and politics are associated with higher user engagement.This large-scale analysis of a complex social and linguistic phenomenon contributes to both NLP and social science research.Frame Type Frame Description Issue-Generic Economic Financial implications of an issue Policy Capacity & Resources The availability or lack of time, physical, human, or financial resources Morality & Ethics Perspectives compelled by religion or secular sense of ethics or social responsibility Fairness & Equality The (in)equality with which laws, punishments, rewards, resources are distributed Legality, Constitutionality & Jurisdiction Court cases and existing laws that regulate policies; constitutional interpretation;

We present an in-depth, language-centric, large-scale study of topical discussions on Twitter by 1,711 Indian politicians from 20 competing political parties in a 4-year timespan. We first show that politicians of all parties collectively indulge more in establishing personal branding through low-substance, personality-focused messaging as opposed to broadcasting policy stances. Additionally, compared to the party-in-power, opposition politicians collectively post more complex tweets and demonstrate higher negativity (e.g., using person-based attack hashtags) especially regarding the issue of corruption. Finally, through a contextual examination of the most retweeted messages from two key leaders---the prime minister and the leader of the largest opposition party, we find that there are qualitatively important distinctions between their styles---while the former focuses on positive-themed messaging, the latter employs confrontation and aggressive language with direct attacks on individuals and issues.

Collaborative crowdsourcing is a well-established model of work, especially in the case of open source software development. The structure and operation of these virtual and loosely-knit teams differ from traditional organizations. As such, little is known about how their behavior may change in response to an increase in external attention. To understand these dynamics, we analyze millions of actions of thousands of contributors in over 1100 open source software projects that topped the GitHub Trending Projects page and thus experienced a large increase in attention, in comparison to a control group of projects identified through propensity score matching. In carrying out our research, we use the lens of organizational change, which considers the challenges teams face during rapid growth and how they adapt their work routines, organizational structure, and management style. We show that trending results in an explosive growth in the effective team size. However, most newcomers make only shallow and transient contributions. In response, the original team transitions towards administrative roles, responding to requests and reviewing work done by newcomers. Projects evolve towards a more distributed coordination model with newcomers becoming more central, albeit in limited ways. Additionally, teams become more modular with subgroups specializing in different aspects of the project. We discuss broader implications for collaborative crowdsourcing teams that face attention shocks.

The spread of content produced by fake news publishers was one of the most discussed characteristics of the 2016 U.S. Presidential Election. Yet, little is known about the prevalence and focus of such content, how its prevalence changed over time, and how this prevalence related to important election dynamics. In this paper, we address these questions using tweets that mention the two presidential candidates sampled at the daily level, the news content mentioned in such tweets, and open-ended responses from nationally representative telephone interviews. The results of our analysis highlight various important lessons for news consumers and journalists. We find that (i.) traditional news producers outperformed fake news producers in aggregate, (ii.) the prevalence of content produced by fake news publishers increased over the course of the campaign-particularly among tweets that mentioned Clinton, and (iii.) changes in such prevalence were closely following changes in net Clinton favorability. Turning to content, we (iv.) identify similarities and differences in agenda setting by fake and traditional news media and show that (v.) information individuals most commonly reported to having read, seen or heard about the candidates was more closely aligned with content produced by fake news outlets than traditional news outlets, in particular for information Republican voters retained about Clinton. We also model fake-ness of retained information as a function of demographics characteristics. Implications for platform owners, news consumers, and journalists are discussed.

Online social movements (OSMs) play a key role in promoting democracy in modern society. Most online activism is largely driven by critical offline events. Among many studies investigating collective behavior in OSMs, few has explored the interaction between crowd dynamics and their offline context. Here, focusing on the Black Lives Matter OSM and utilizing an event-driven approach on a dataset of 36 million tweets and thousands of offline events, we study how different types of offline events-police violence and heightened protests-influence crowd behavior over time. We find that police violence events and protests play important roles in the recruitment process. Moreover, by analyzing the re-participation dynamics and patterns of social interactions, we find that, in the long term, users who joined the movement during police violence events and protests show significantly more commitment than those who joined during other times. However, users recruited during other times are more committed to the movement than the other two groups in the short term. Furthermore, we observe that social ties formed during police violence events are more likely to be sustained over time than those formed during other times. Contrarily, ties formed during protests are the least likely to be maintained. Altogether, our results shed light on the impact of bursting events on the recruitment, retention, and communication patterns of collective behavior in the Black Lives Matter OSM.

This paper explores Community-Based Trip Sharing which uses the structure of communities and commuting patterns to optimize car or ride sharing for urban communities. It introduces the Commuting Trip Sharing Problem (CTSP) and proposes an optimization approach to maximize trip sharing. The optimization method, which exploits trip clustering, shareability graphs, and mixed-integer programming, is applied to a dataset of 9000 daily commuting trips from a mid-size city. Experimental results show that community-based trip sharing reduces daily car usage by up to 44%, thus producing significant environmental and traffic benefits and reducing parking pressure. The results also indicate that daily flexibility in pairing cars and passengers has significant impact on the benefits of the approach, revealing new insights on commuting patterns and trip sharing.

The Guardian ---the fifth most widely read online newspaper in the world as of 2014---changed conversations on its commenting platform by altering its design from non-threaded to single-level threaded in 2012. We studied this naturally occurring experiment to investigate the impact of conversation threading on user retention as mediated by several potential changes in conversation structure and style. Our analysis shows that the design change made new users significantly more likely to comment a second time, and that this increased stickiness is due in part to a higher fraction of comments receiving responses after the design change. In mediation analysis, other anticipated mechanisms such as reciprocal exchanges and comment civility did not help to explain users' decision to return to the commenting system; indeed, civility did not increase after the design change and reciprocity declined. These analyses show that even simple design choices can have a significant impact on news forums' stickiness. Further, they suggest that this influence is more powerfully shaped by affordances---the new system made responding easier---than by changes in users' attention to social norms of reciprocity or civility. This has an array of implications for designers.

Collaborative crowdsourcing is an increasingly common way of accomplishing work in our economy. Yet, we know very little about how the behavior of these crowds changes over time and how these dynamics impact their performance. In this paper, we take a group development approach that considers how the behavior of crowds change over time in anticipation and as a result of their evaluation and recognition. Towards this goal, this paper studies the collaborative behavior of groups comprised of editors of articles that have been recognized for their outstanding quality and given the Good Articles (GA) status and those that eventually become Featured Articles (FA) on Wikipedia. The results show that the collaborative behavior of GA groups radically changes just prior to their nomination. In particular, the GA groups experience increases in the level of activity, centralization of workload, and level of GA experience and decreases in conflict (i.e., reverts) among editors. After being promoted to GA, they converge back to their typical behavior and composition. This indicates that crowd behavior prior to their evaluation period is dramatically different than behavior before or after. In addition, the collaborative behaviors of crowds during their promotion to GA are predictive of whether they are eventually promoted to FA. Our findings shed new light on the importance of time in understanding the relationship between crowd performance and collaborative measures such as centralization, conflict and experience.

Two recent disruptions to the online advertising market are the widespread use of ad-blocking software and proposed restrictions on third-party tracking, trends that are driven largely by consumer concerns over privacy. Both primarily impact display advertising (as opposed to search and native social ads), and affect how retailers reach customers and how content producers earn revenue. It is, however, unclear what the consequences of these trends are. We investigate using anonymized web browsing histories of 14 million individuals, focusing on "retail sessions" in which users visit online sites that sell goods and services. We find that only 3% of retail sessions are initiated by display ads, a figure that is robust to permissive attribution rules and consistent across widely varying market segments. We further estimate the full distribution of how retail sessions are initiated, and find that search advertising is three times more important than display advertising to retailers, and search advertising is itself roughly three times less important than organic web search. Moving to content providers, we find that display ads are shown by 12% of websites, accounting for 32% of their page views; this reliance is concentrated in online publishing, e.g., news outlets) where the rate is 91%. While most consumption is either in the long-tail of websites that do not show ads, or sites like Facebook that show native, first-party ads, moderately sized web publishers account for a substantial fraction of consumption, and we argue that they will be most affected by changes in the display advertising market. Finally, we use estimates of ad rates to judge the feasibility of replacing lost ad revenue with a freemium or donation-based model.

The success of a group depends on continued participation of its members through time. We study the factors that affect continued user participation in the context of educational Twitter chats. To predict whether a user that attended her first session in a particular Twitter chat group will return to the group, we build 5F Model that captures five different factors: individual initiative, group characteristics, perceived receptivity, linguistic affinity and geographical proximity. Through statistical data analysis of thirty Twitter chats over a two year period as well as a survey study, our work provides many insights about group dynamics in Twitter chats. We show similarities between Twitter chats and traditional groups such as the importance of social inclusion and linguistic similarity while also identifying important distinctions such as the insignificance of geographical proximity. We also show that informational support is more important than emotional support in educational Twitter chats, but this does not reduce the sense of community as suggested in earlier studies.

The First Law of Geography states "Everything is related to everything else, but near things are more related than distant things". This spatial significance has implications in various applications, trend detection being one of them. In this paper we propose a new algorithmic tool, GeoScope , to detect geo-trends. GeoScope is a data streams solution that detects correlations between topics and locations in a sliding window, in addition to analyzing topics and locations independently. GeoScope offers theoretical guarantees for detecting all trending correlated pairs while requiring only sub-linear space and running time. We perform various human validation tasks to demonstrate the value of GeoScope. The results show that human judges prefer GeoScope to the best performing baseline solution 4:1 in terms of the geographical significance of the presented information. As the Twitter analysis demonstrates, GeoScope successfully filters out topics without geo-intent and detects various local interests such as emergency events, political demonstrations or cultural events. Experiments on Twitter show that GeoScope has perfect recall and near-perfect precision.

Recent studies on the diffusion of information in social networks have largely focused on models based on the influence of local friends. In this paper, we challenge the generalizability of this approach and revive theories introduced by social scientists in the context of diffusion of innovations to model user behavior. To this end, we study various diffusion models in two different online social networks, Digg and Twitter. We first evaluate the applicability of two representative local influence models and show that the behavior of most social networks users are not captured by these local models. Next, driven by theories introduced in the diffusion of innovations research, we introduce a novel diffusion model called Gaussian Logit Curve Model (GLCM) that models user behavior with respect to the behavior of the general population. Our analysis shows that GLCM captures user behavior significantly better than local models, especially in the context of Digg. Aiming to capture both the local and global signals, we introduce various hybrid models and evaluate them through statistical methods. Our methodology models each user separately, automatically determining which users are driven by their local relations and which users are better defined through adopter categories, therefore capturing the complexity of human behavior.

Information diffusion in social networks provide great opportunities for political and social change as well as societal education. Therefore understanding information diffusion in social networks is a critical research goal. This greater understanding can be achieved through data analysis, development of reliable models that can predict outcomes of social processes, and ultimately the creation of applications that can shape the outcome of these processes. In this tutorial, we aim to provide an overview of such recent research based on a wide variety of techniques such as optimization algorithms, data mining, data streams covering a large number of problems such as influence spread maximization, misinformation limitation and study of trends in online social networks.

In this work, we study the notion of competing campaigns in a social network and address the problem of influence limitation where a "bad" campaign starts propagating from a certain node in the network and use the notion of limiting campaigns to counteract the effect of misinformation. The problem can be summarized as identifying a subset of individuals that need to be convinced to adopt the competing (or "good") campaign so as to minimize the number of people that adopt the "bad" campaign at the end of both propagation processes. We show that this optimization problem is NP-hard and provide approximation guarantees for a greedy solution for various definitions of this problem by proving that they are submodular. We experimentally compare the performance of the greedy method to various heuristics. The experiments reveal that in most cases inexpensive heuristics such as degree centrality compare well with the greedy approach. We also study the influence limitation problem in the presence of missing data where the current states of nodes in the network are only known with a certain probability and show that prediction in this setting is a supermodular problem. We propose a prediction algorithm that is based on generating random spanning trees and evaluate the performance of this approach. The experiments reveal that using the prediction algorithm, we are able to tolerate about 90% missing data before the performance of the algorithm starts degrading and even with large amounts of missing data the performance degrades only to 75% of the performance that would be achieved with complete data.

The identification of popular and important topics discussed in social networks is crucial for a better understanding of societal concerns. It is also useful for users to stay on top of trends without having to sift through vast amounts of shared information. Trend detection methods introduced so far have not used the network topology and has thus not been able to distinguish viral topics from topics that are diffused mostly through the news media. To address this gap, we propose two novel structural trend definitions we call coordinated and uncoordinated trends that use friendship information to identify topics that are discussed among clustered and distributed users respectively. Our analyses and experiments show that structural trends are significantly different from traditional trends and provide new insights into the way people share information online. We also propose a sampling technique for structural trend detection and prove that the solution yields in a gain in efficiency and is within an acceptable error bound. Experiments performed on a Twitter data set of 41.7 million nodes and 417 million posts show that even with a sampling rate of 0.005, the average precision is 0.93 for coordinated trends and 1 for uncoordinated trends.