Search

Opinion dynamics are affected by cognitive biases and noise. While mathematical models have focused extensively on biases, we still know surprisingly little about how noise shapes opinion patterns. Here, we use an agent-based opinion dynamics model to investigate the interplay between confirmation bias-represented as bounded confidence-and different types of noise. After analysing where noise can enter social interaction, we propose a type of noise that has not been discussed so far, ambiguity noise. While previously considered types of noise acted on agents either before, after or independent of social interaction, ambiguity noise acts on communicated messages, assuming that socially transmitted opinions are inherently noisy. We find that noise can induce agreement when confirmation bias is moderate, but different types of noise require quite different conditions for this effect to occur. An application of our model to the climate change debate shows that at just the right mix of confirmation bias and ambiguity noise, opinions tend to converge to high levels of climate change concern. This result is not observed with the other types. Our findings highlight the importance of considering and distinguishing between the various types of noise and the unique role of ambiguity in opinion formation.

America is said to be more polarized than ever before, and extensive research examines the causes and effects of political polarization. A less studied but more pronounced trend is that citizens perceive society and politics as sharply divided. We focus on news exposure as a key reason for these perceptions. We disentangle the unique effects of exposure to partisan outlets and to news coverage of polarization in general, whether in partisan or centrist news media. We rely on 9 months of online behavioral data (127,400,000 visits) from large samples in the U.S. and Poland (total N = 2,462 & 2,120), paired with a three-wave panel survey on participants' attitudes and perceptions. We measure individual exposure, over-time and unobtrusively, to partisan outlets as well as to news coverage of polarization through a fine-tuned BERT-based machine learning classifier. We find that both exposures increase perceived polarization, independently, in both countries. Yet, in both contexts, it is news coverage of polarization, regardless of whether it is ideologically left, right, or centrist news outlets, that is the strongest driver of perceived polarization. These changes are confined to personal-level polarization (perceived distance between oneself and members of the opposing party), while no measure significantly influences perceptions of society, more generally, as being polarized. These findings have important implications for our understanding of how the media’s increasing tendency to focus on conflict as a central reporting frame can be the driver of often exaggerated perceptions of partisan divisions. [ABSTRACT FROM AUTHOR]

The issue of opinion sharing and formation has received considerable attention in the academic literature, and a few models have been proposed to study this problem. However, existing models are limited to the interactions among nearest neighbors, with those second, third, and higher-order neighbors only considered indirectly, despite the fact that higher-order interactions occur frequently in real social networks. In this paper, we develop a new model for opinion dynamics by incorporating long-range interactions based on higher-order random walks that can explicitly tune the degree of influence of higher-order neighbor interactions. We prove that the model converges to a fixed opinion vector, which may differ greatly from those models without higher-order interactions. Since direct computation of the equilibrium opinion is computationally expensive, which involves the operations of huge-scale matrix multiplication and inversion, we design a theoretically convergence-guaranteed estimation algorithm that approximates the equilibrium opinion vector nearly linearly in both space and time with respect to the number of edges in the graph. We conduct extensive experiments on various social networks, demonstrating that the new algorithm is both highly efficient and effective. [ABSTRACT FROM AUTHOR]

Scientific datasets play a crucial role in contemporary data-driven research, as they allow for the progress of science by facilitating the discovery of new patterns and phenomena. This mounting demand for empirical research raises important questions on how strategic data utilization in research projects can stimulate scientific advancement. In this study, we examine the hypothesis inspired by the recombination theory, which suggests that innovative combinations of existing knowledge, including the use of unusual combinations of datasets, can lead to high-impact discoveries. Focusing on social science, we investigate the scientific outcomes of such atypical data combinations in more than 30,000 publications that leverage over 5,000 datasets curated within one of the largest social science databases, Interuniversity Consortium for Political and Social Research. This study offers four important insights. First, combining datasets, particularly those infrequently paired, significantly contributes to both scientific and broader impacts (e.g., dissemination to the general public). Second, infrequently paired datasets maintain a strong association with citation even after controlling for the atypicality of dataset topics. In contrast, the atypicality of dataset topics has a much smaller positive impact on citation counts. Third, smaller and less experienced research teams tend to use atypical combinations of datasets in research more frequently than their larger and more experienced counterparts. Last, despite the benefits of data combination, papers that amalgamate data remain infrequent. This finding suggests that the unconventional combination of datasets is an underutilized but powerful strategy correlated with the scientific impact and broader dissemination of scientific discoveries. [ABSTRACT FROM AUTHOR]

Automated social media accounts, known as bots, are increasingly recognized as key tools for manipulative online activities. These activities can stem from coordination among several accounts and these automated campaigns can manipulate social network structure by following other accounts, amplifying their content, and posting messages to spam online discourse. In this study, we present a novel unsupervised detection method designed to target a specific category of malicious accounts designed to manipulate user metrics such as online popularity. Our framework identifies anomalous following patterns among all the followers of a social media account. Through the analysis of a large number of accounts on the Twitter platform (rebranded as X after the acquisition of Elon Musk), we demonstrated that irregular following patterns are prevalent and are indicative of automated fake accounts. Notably, we found that these detected groups of anomalous followers exhibited consistent behavior across multiple accounts. This observation, combined with the computational efficiency of our proposed approach, makes it a valuable tool for investigating large-scale coordinated manipulation campaigns on social media platforms. [ABSTRACT FROM AUTHOR]

Carefully standardized facial images of 591 participants were taken in the laboratory while controlling for self-presentation, facial expression, head orientation, and image properties. They were presented to human raters and a facial recognition algorithm: both humans (r =.21) and the algorithm (r =.22) could predict participants' scores on a political orientation scale (Cronbach's α =.94) decorrelated with age, gender, and ethnicity. These effects are on par with how well job interviews predict job success, or alcohol drives aggressiveness. The algorithm's predictive accuracy was even higher (r =.31) when it leveraged information on participants' age, gender, and ethnicity. Moreover, the associations between facial appearance and political orientation seem to generalize beyond our sample: The predictive model derived from standardized images (while controlling for age, gender, and ethnicity) could predict political orientation (r ≈.13) from naturalistic images of 3,401 politicians from the United States, the United Kingdom, and Canada. The analysis of facial features associated with political orientation revealed that conservatives tended to have larger lower faces. The predictability of political orientation from standardized images has critical implications for privacy, the regulation of facial recognition technology, and understanding the origins and consequences of political orientation. Public Significance Statement: We demonstrate that political orientation can be predicted from neutral facial images by both humans and algorithms, even when factors like age, gender, and ethnicity are accounted for. This indicates a connection between political leanings and inherent facial characteristics, which are largely beyond an individual's control. Our findings underscore the urgency for scholars, the public, and policymakers to recognize and address the potential risks of facial recognition technology to personal privacy. [ABSTRACT FROM AUTHOR]

Scholars have access to a rich source of political discourse via social media. Although computational approaches to understand this communication are being used, they tend to be unsupervised and off-the-shelf algorithms to describe a corpus of messages. This article details our approach at using human-supervised machine learning to study political campaign messages. Although some declare this technique too labor-intensive, it provides theoretically informed classification, making it more accurate and reliable. This article describes the design decisions and accuracy of our algorithms, and the applicability of the approach to classifying messages from Facebook and Twitter across two cultures and to advertisements. [ABSTRACT FROM AUTHOR]

We investigated large language models' (LLMs) efficacy in classifying complex psychological constructs like intellectual humility, perspective-taking, open-mindedness, and search for a compromise in narratives of 347 Canadian and American adults reflecting on a workplace conflict. Using state-of-the-art models like GPT-4 across few-shot and zero-shot paradigms and RoB-ELoC (RoBERTa -fine-tuned-on-Emotion-with-Logistic-Regression-Classifier), we compared their performance with expert human coders. Results showed robust classification by LLMs, with over 80% agreement and F1 scores above 0.85, and high human-model reliability (Cohen's κ Md across top models =.80). RoB-ELoC and few-shot GPT-4 were standout classifiers, although somewhat less effective in categorizing intellectual humility. We offer example workflows for easy integration into research. Our proof-of-concept findings indicate the viability of both open-source and commercial LLMs in automating the coding of complex constructs, potentially transforming social science research. [ABSTRACT FROM AUTHOR]

Although journalists' social media sourcing can empower non-elite sources and diversify public discussions, counterarguments maintain that social media sourcing relies on a small group of elites and reinforces social division. To contribute to that debate, we examined how health journalists from the mainstream news organizations in the U.S. used Twitter's @mention for sourcing during the first three months of the COVID-19 outbreak. Using a sample of public Twitter posts published by the journalists, we formed co-@mentioned networks (i.e., two sources were connected if @mentioned in the same post) to examine the structure of the networks and identify important sourcing informants. Among the results, elite sources (e.g., health journalists and health experts in the public sector) and influential users (i.e., verified users with a large number of followers and who post frequently) dominated the sourcing repertoire. Moreover, the networks were fragmented because the sources were clustered into several close-knit subgroups. Analyzing exponential random graph models to examine the formation mechanism of the networks revealed that, as the pandemic's severity increased, influential users played a more salient role in the sourcing repertoire, and a homogeneous cluster consisting of journalists and news organizations emerged. [ABSTRACT FROM AUTHOR]

Social influence aligns peoples opinions, but social identities and related in-group biases interfere with this alignment. For instance, the recent rise of young climate activists (e.g. Fridays for Future or Last Generation) has highlighted the importance of generational identities in the climate change debate. It is unclear how social identities affect the emergence of opinion patterns, such as consensus or disagreement, in a society. Here, we present an agent-based model to explore this question. Agents communicate in a network and form opinions through social influence. The agents have fixed social identities which involve homophily in their interaction preferences and in-group bias in their perception of others. We find that the in-group bias has opposing effects depending on the network topology. The bias impedes consensus in highly random networks by promoting the formation of echo chambers within social identity groups. By contrast, the bias facilitates consensus in highly clustered networks by aligning dispersed in-group agents across the network and, thereby, preventing the formation of isolated echo chambers. Our model uncovers the mechanisms underpinning these opposing effects of the in-group bias and highlights the importance of the communication network topology for shaping opinion dynamics.

Abstract The progress in high-performance computing makes it increasingly possible to build detailed models to investigate spreading processes on complex networks. However, current studies have been lacking detailed computational methods to describe spreading processes in large complex networks. To fill this gap we present a new modelling approach for analysing influence spreading via individual nodes and links on various network structures. The proposed influence-spreading model uses a probability matrix to capture the spreading probability from one node to another in the network. This approach enables analysing network characteristics in a number of applications and spreading processes using metrics that are consistent with the quantities used to model the network structures. In addition, this study combines sub-models and offers a comprehensive look at different applications and metrics previously discussed in cases of social networks, community detection, and epidemic spreading. Here, we also note that the centrality measures based on the probability matrix are used to identify the most significant nodes in the network. Furthermore, the model can be expanded to include additional properties, such as introducing individual breakthrough probabilities for the nodes and specific temporal distributions for the links.

Abstract Automated social media accounts, known as bots, are increasingly recognized as key tools for manipulative online activities. These activities can stem from coordination among several accounts and these automated campaigns can manipulate social network structure by following other accounts, amplifying their content, and posting messages to spam online discourse. In this study, we present a novel unsupervised detection method designed to target a specific category of malicious accounts designed to manipulate user metrics such as online popularity. Our framework identifies anomalous following patterns among all the followers of a social media account. Through the analysis of a large number of accounts on the Twitter platform (rebranded as X after the acquisition of Elon Musk), we demonstrated that irregular following patterns are prevalent and are indicative of automated fake accounts. Notably, we found that these detected groups of anomalous followers exhibited consistent behavior across multiple accounts. This observation, combined with the computational efficiency of our proposed approach, makes it a valuable tool for investigating large-scale coordinated manipulation campaigns on social media platforms.

Abstract Large language models have become popular over a short period of time because they can generate text that resembles human writing across various domains and tasks. The popularity and breadth of use also put this technology in the position to fundamentally reshape how written language is perceived and evaluated. It is also the case that spoken language has long played a role in maintaining power and hegemony in society, especially through ideas of social identity and “correct” forms of language. But as human communication becomes even more reliant on text and writing, it is important to understand how these processes might shift and who is more likely to see their writing styles reflected back at them through modern AI. We therefore ask the following question: who does generative AI write like? To answer this, we compare writing style features in over 150,000 college admissions essays submitted to a large public university system and an engineering program at an elite private university with a corpus of over 25,000 essays generated with GPT-3.5 and GPT-4 to the same writing prompts. We find that human-authored essays exhibit more variability across various individual writing style features (e.g., verb usage) than AI-generated essays. Overall, we find that the AI-generated essays are most similar to essays authored by students who are males with higher levels of social privilege. These findings demonstrate critical misalignments between human and AI authorship characteristics, which may affect the evaluation of writing and calls for research on control strategies to improve alignment.

Collective intelligence challenges are often entangled with collective action problems. For example, voting, rating, and social innovation are collective intelligence tasks that require costly individual contributions. As a result, members of a group often free ride on the information contributed by intrinsically motivated people. Are intrinsically motivated agents the best participants in collective decisions? We embedded a collective intelligence task in a large-scale, virtual world public good game and found that participants who joined the information system but were reluctant to contribute to the public good (free riders) provided more accurate evaluations, whereas participants who rated frequently underperformed. Testing the underlying mechanism revealed that a negative rating bias in free riders is associated with higher accuracy. Importantly, incentivizing evaluations amplifies the relative influence of participants who tend to free ride without altering the (higher) quality of their evaluations, thereby improving collective intelligence. These results suggest that many of the currently available information systems, which strongly select for intrinsically motivated participants, underperform and that collective intelligence can benefit from incentivizing free riding members to engage. More generally, enhancing the diversity of contributor motivations can improve collective intelligence in settings that are entangled with collective action problems.

This article explores the role of hackathons for good in building a community of software and hardware developers focused on addressing global sustainable development goal (SDG) challenges. We theorize this movement as computational diplomacy: a decentralized, participatory process for digital governance that leverages collective intelligence to tackle major global issues. Analysing Devpost and GitHub data reveals that 30% of hackathons since 2010 have addressed SDG topics, employing diverse technologies to create innovative solutions. Hackathons serve as crucial kairos moments, sparking innovation bursts that drive both immediate project outcomes and long-term production. We propose that these events harness the neurobiological basis of human cooperation and empathy, fostering a collective sense of purpose and reducing interpersonal prejudice. This bottom–up approach to digital governance integrates software development, human collective intelligence and collective action, creating a dynamic model for transformative change. By leveraging kairos moments, computational diplomacy promotes a more inclusive and effective model for digital multilateral governance of the future. This article is part of the theme issue 'Co-creating the future: participatory cities and digital governance'. [ABSTRACT FROM AUTHOR]

Technological artefacts are created in accordance with the values and worldviews of their designers. In operation, they act as a medium, facilitating and constraining human interaction with, and perception of, the world. When used on a large scale, they may lastingly affect societal ethos. If institutional structures of domination allocate the resources necessary for artefact design and development to some population groups over others, the direction and extent of such an effect may lead to increased disparity and inequity. While the direct influence of technology on opinion is well-studied, the evaluation of non-epistemic values, assumptions and presuppositions is a hurdle in the way of a deeper understanding of the large-scale effects of asymmetries in worldviews embodied by artefacts. Here, we show that artefacts have a strong potential to bias societal worldviews when they are distributed unevenly across the value spectrum. They can affect the clustering behaviour of agents with regard to worldview, both aiding and hindering intra- and inter-cluster diversity, depending on their distribution and frequency. Our findings underline the distributional sensitivity of worldview dynamics to institutional structures of domination. We highlight the importance of procedural interventions such as participatory design, which explicitly acknowledges existing asymmetries and redistributes power accordingly. This article is part of the theme issue 'Co-creating the future: participatory cities and digital governance'. [ABSTRACT FROM AUTHOR]

The use of individual-level browsing data, that is, the records of a person's visits to online content through a desktop or mobile browser, is of increasing importance for social scientists. Browsing data have characteristics that raise many questions for statistical analysis, yet to date, little hands-on guidance on how to handle them exists. Reviewing extant research, and exploring data sets collected by our four research teams spanning seven countries and several years, with over 14,000 participants and 360 million web visits, we derive recommendations along four steps: preprocessing the raw data; filtering out observations; classifying web visits; and modelling browsing behavior. The recommendations we formulate aim to foster best practices in the field, which so far has paid little attention to justifying the many decisions researchers need to take when analyzing web browsing data. [ABSTRACT FROM AUTHOR]

Abstract Social media manipulation poses a significant threat to cognitive autonomy and unbiased opinion formation. Prior literature explored the relationship between online activity and emotional state, cognitive resources, sunlight and weather. However, a limited understanding exists regarding the role of time of day in content spread and the impact of user activity patterns on susceptibility to mis- and disinformation. This work uncovers a strong correlation between user activity time patterns and the tendency to spread potentially disinformative content. Through quantitative analysis of Twitter (now X) data, we examine how user activity throughout the day aligns with diurnal behavioural archetypes. Evening types exhibit a significantly higher inclination towards spreading potentially disinformative content, which is more likely at night-time. This knowledge can become crucial for developing targeted interventions and strategies that mitigate misinformation spread by addressing vulnerable periods and user groups more susceptible to manipulation.

Language shapes diverse cultures and creates natural barriers between human societies. The landscape of nonprofit and philanthropic studies in non-English languages is barely charted, impeding the globalization of this research field. This project (a) describes the topics shared between English and Chinese scholarship on nonprofits and philanthropy and (b) explores why English scholarship is cited in Chinese journal articles from five aspects: rationale of scholarship, novelty, relevance, social network, and reputation. The English articles cited by Chinese scholars tend to: (a) focus on instrumentality but not expressive values, (b) develop rather than disrupt existing paradigms, and (c) be relevant to topics popular in Chinese literature and have authors with Chinese scholarly connections. In general, Chinese scholars tend to cite English articles that are value-neutral, non-disruptive, and native. Theoretical and methodological implications for examining nonprofit studies in other languages are discussed. [ABSTRACT FROM AUTHOR]

The recent so-called 'data revolution' offers unprecedented opportunities to innovate regional policies. New data sources are being widely used by the scientific community, however their uptake is far from being systematic in the policy cycle, where data innovation can improve territorial impact assessment. This paper presents a survey on the use of non-traditional data in the context of regional policy, together with a case study on real estate markets of three European countries, highlighting the perspectives and limitations of computational social science in regional analysis in terms of data quality and availability. [ABSTRACT FROM AUTHOR]

A significant advancement in marketing studies has been the use of big data analytics to explore and derive customer experience (CX) insights. With fitness businesses' pivoting towards managing CX along the entire customer journey, this study aimed to understand CX in the commercial fitness industry. A textual corpus of three million words was extracted from a total of 24,231 Yelp reviews on 1,045 fitness clubs in the United States. Structural Topic Modeling (STM), content analysis, and topic network analysis were performed to identify CX themes and their inter-relationships. We then inductively identified topics and themes germane to CX in fitness clubs and derived a general CX model. Sixty-eight relevant topics were identified through an iterative STM and content analysis process and were subsequently interpreted using the CX paradigm and experience design framework (Funk, 2017; Voss et al., 2008). Our final model integrates experience design elements (i.e. stageware, orgware, customerware), customer journey (i.e. pre-usage, usage, post-usage) and CX (i.e. subjective and internal responses) and shows the interrelationships among them. Our model offers an integrative and coherent theoretical framework to examine CX in commercial fitness clubs. Our model and methodology allow researchers and organizations not only to capture CX and the relative salience of CX topics, but also to track CX trends over time, compare the experience design elements of an organization, measure CX performance against competitors, and link CX topics to other criterion variables (e.g. retention rate, financial performance, business survivability). [ABSTRACT FROM AUTHOR]

The rise of social media as a source of science and health information has brought challenges to informed citizenship and social trust due to the spread of misinformation, particularly anti-vaccination messages that incite hatred and discourage necessary health precautions. These messages often employ emotional appeals and identity cues. However, scholarship examining emotional appeals and identity cues in anti-vax messages is still at the nascent stage. Furthermore, most literature on emotions and identities on social media has focused on text-based platforms, despite the increasing popularity of interactive, multimodal platforms. To address these gaps, our paper analyzes recent TikTok anti-vax videos and incorporates the framework of multimodal frame processing, emotion-as-frames model, affective intelligence theory, and social identity theory. Our paper uncovers how different message modalities affect the impact of emotional narratives and identity cues on user engagement. We also investigate sociopolitical identity cues beyond partisan identities, expanding the current terrain of political communication. Our results demonstrate that audiences engage with emotional and identity cues in anti-vax videos differently based on distinct message modalities. We also found that identity cues related to interpersonal relationships (e.g. parental) and conspiracy groups were prevalent, in addition to partisan identity cues. These results offer new insights into sociopolitical identities beyond partisanship and highlight the importance of considering the multi-modal nature of video platforms. Overall, our paper sheds light on the complex relationship between emotions, identities, and message modalities on social media and provides important implications for addressing misinformation and improving science communication on digital platforms. [ABSTRACT FROM AUTHOR]

Contention over COVID-19 is only a recent example of increasing social division around science in the U.S. Many blame these divisions on actors who have strategically sowed doubt and distrust around expert supported positions and policies. However, this overlooks how scientists have fueled narratives of social and political conflict around science. This study explores how science influencers on social media have used group identity language in ways that may perpetuate narratives of intergroup conflict around science. Using computer-assisted content analytic methods, we examine how science influencers' use of group identity language has changed in response to recent events (Trump presidency, COVID-19 pandemic) and across different social media platforms (Twitter, Facebook, Instagram). While there are slight increases in group identity language between 2016 and 2021, different patterns across platforms suggest that science influencers use different platforms to perform multiple roles of engaging diverse audiences, building ingroup solidarity, and defending against outgroup criticism. [ABSTRACT FROM AUTHOR]

Among many by-products of Web 2.0 come the wide range of potential image and text datasets within social media and content sharing platforms that speak of how people live, what they do, and what they care about. These datasets are imperfect and biased in many ways, but those flaws make them complementary to data derived from conventional social science methods and thus potentially useful for triangulation in complex decision-making contexts. Yet the online environment is highly mutable, and so the datasets are less reliable than censuses or other standard data types leveraged in social impact assessment. Over the past decade, we have innovated numerous methods for deploying Instagram datasets in investigating management or development alternatives. This article synthesizes work from three Canadian decision contexts – hydroelectric dam construction or removal; dyke realignment or wetland restoration; and integrating renewable energy into vineyard landscapes – to illustrate some of the methods we have applied to social impact assessment questions using Instagram that may be transferrable to other social media platforms and contexts: thematic (manual coding, machine vision, natural language processing/sentiment analysis, statistical analysis), spatial (hotspot mapping, cultural ecosystem modeling), and visual (word clouds, saliency mapping, collage). We conclude with a set of cautions and next steps for the domain. [ABSTRACT FROM AUTHOR]

Recital 33 GDPR has often been interpreted as referring to 'broad consent'. This version of informed consent was intended to allow data subjects to provide their consent for certain areas of research, or parts of research projects, conditional to the research being in line with 'recognised ethical standards'. In this article, we argue that broad consent is applicable in the emerging field of Computational Social Science (CSS), which lies at the intersection of data science and social science. However, the lack of recognised ethical standards specific to CSS poses a practical barrier to the use of broad consent in this field and other fields that lack recognised ethical standards. Upon examining existing research ethics standards in social science and data science, we argue that they are insufficient for CSS. We further contend that the fragmentation of European Union (EU) law and research ethics sources makes it challenging to establish universally recognised ethical standards for scientific research. As a result, CSS researchers and other researchers in emerging fields that lack recognised ethical standards are left without sufficient guidance on the use of broad consent as provided for in the GDPR. We conclude that responsible EU bodies should provide additional guidance to facilitate the use of broad consent in CSS research. [ABSTRACT FROM AUTHOR]

This article provides a sociology of Twitter (now known as X) and charts the development of the study of the platform and its data in sociological venues through an analysis of 1,644 articles published since 2009. This review helps readers understand developments in the field and provides a road map for advancing future Twitter-related sociological research. The works cited in this review advance sociological research on a variety of subfields, including but not limited to race, social movements, segregation, politics, violence, and stratification. After a brief introduction of definitions as well as methodological approaches used to study the platform (computational, qualitative, and mixed), I explore how Twitter has been used in sociological research. Furthermore, using social movements and activism as a case study, I highlight what the platform's usage and communication reveal about the social world. I conclude with a road map for advancing Twitter-related sociological research in the current atmosphere in which Twitter has been renamed X. Ultimately, a sociology of Twitter does not need to be tethered to the platform per se, as it also provides a framework for understanding new platforms if they become home for what people previously posted to Twitter/X. [ABSTRACT FROM AUTHOR]

In this paper, we introduce ProxMetrics, a novel toolkit designed to evaluate similarity among social media entities through proxemic dimensions. Proxemics is the science that studies the organization of space and the effects of distances on behavior and interactions. It encompasses 5 core dimensions: Distance, Identity, Location, Movement, and Orientation. Adapting the principles of traditional physical proxemics to the digital world of social media, we present a method and a modular similarity function to determine proxemic similarity scores across heterogeneous social media entities (users, groups, places, themes and times) based on these dimensions. The approach used is intended to be modular and generic, ensuring adaptability across various application domains and requirements. The calculated scores act as indicators and offer valuable insights for stakeholders, aligning with distinct domain requirements. Empirical testing in the tourism domain highlights the toolkit's extensive applicability across a variety of requirements. [ABSTRACT FROM AUTHOR]

For qualitative data analysis (QDA), researchers assign codes to text segments to arrange the information into topics or concepts. These annotations facilitate information retrieval and the identification of emerging patterns in unstructured data. However, this metadata is typically not published or reused after the research. Subsequent studies with similar research questions require a new definition of codes and do not benefit from other analysts' experience. Machine learning (ML) based classification seeded with such data remains a challenging task due to the ambiguity of code definitions and the inherent subjectivity of the exercise. Previous attempts to support QDA using ML rely on linear models and only examined individual datasets that were either smaller or coded specifically for this purpose. However, we show that modern approaches effectively capture at least part of the codes' semantics and may generalize to multiple studies. We analyze the performance of multiple classifiers across three large real-world datasets. Furthermore, we propose an ML-based approach to identify semantic relations of codes in different studies to show thematic faceting, enhance retrieval of related content, or bootstrap the coding process. These are encouraging results that suggest how analysts might benefit from prior interpretation efforts, potentially yielding new insights into qualitative data. [ABSTRACT FROM AUTHOR]

In the growing trend of research using digital trace data to study human activities and opinions across different contexts, networked China has emerged as a prominent area of interest. However, research that critically examines the use, strengths, and weaknesses of existing digital trace methods, and the extent to which they can reveal the true landscape of digital China remains limited. To address these gaps, this study proposes a framework for examining and evaluating the knowledge production of digital trace research within a sociotechnical system comprising state actors, platform governance, digital civil society, and international forces. We then provide the first empirical examination of the knowledge claims and epistemic approaches used in digital trace communication scholarship that has studied China across different phases in the past 30 years. Grounded in the resulting empirical evidence, we discuss two common practices in existing digital trace research on China, how these approaches and perspectives could affect the validity and reliability of offering diverse viewpoints for studying and understanding digital China, and directions for improving these practices. [ABSTRACT FROM AUTHOR]

Despite the growing capabilities of large language models, concerns exist about the biases they develop. In this paper, we propose a novel, automated mechanism for debiasing through specified dataset augmentation in the lens of bias producers that can be useful in a variety of industries, especially ones that are "restricted" and have limited data. We consider that bias can occur due to intrinsic model architecture and dataset quality. The two aspects are evaluated using two different metrics we created. We show that our dataset augmentation algorithm reduces bias as measured by our metrics. Our code can be found on an online GitHub repository. [ABSTRACT FROM AUTHOR]

The paper addresses some fundamental and hotly debated issues for high-stakes event predictions underpinning the computational approach to social sciences, especially in criminology and criminal justice. We question several prevalent views against machine learning and outline a new paradigm that highlights the promises and promotes the infusion of computational methods and conventional social science approaches. [ABSTRACT FROM AUTHOR]

Abstract The same individuals can express very different emotions in online social media with respect to face-to-face interactions, partially because of intrinsic limitations of the digital environments and partially because of their algorithmic design, which is optimized to maximize engagement. Such differences become even more pronounced for topics concerning socially sensitive and polarizing issues, such as massive pharmaceutical interventions. Here, we investigate how online emotional responses change during the large-scale COVID-19 vaccination campaign with respect to a baseline in which no specific contentious topic dominates. We show that the online discussions during the pandemic generate a vast spectrum of emotional response compared to the baseline, especially when we take into account the characteristics of the users and the type of information shared in the online platform. Furthermore, we analyze the role of the political orientation of shared news, whose circulation seems to be driven not only by their actual informational content but also by the social need to strengthen one’s affiliation to, and positioning within, a specific online community by means of emotionally arousing posts. Our findings stress the importance of better understanding the emotional reactions to contentious topics at scale from digital signatures, while providing a more quantitative assessment of the ongoing online social dynamics to build a faithful picture of offline social implications.

Abstract In this paper, we engage with and expand on the keynote talk about the “Right to Audit” given by Prof. Christian Sandvig at the International Conference on Computational Social Science 2021 through a critical reflection on power asymmetries in the algorithm auditing field. We elaborate on the challenges and asymmetries mentioned by Sandvig — such as those related to legal issues and the disparity between early-career and senior researchers. We also contribute a discussion of the asymmetries that were not covered by Sandvig but that we find critically important: those related to other disparities between researchers, incentive structures related to the access to data from companies, targets of auditing and users and their rights. We also discuss the implications these asymmetries have for algorithm auditing research such as the Western-centrism and the lack of the diversity of perspectives. While we focus on the field of algorithm auditing specifically, we suggest some of the discussed asymmetries affect Computational Social Science more generally and need to be reflected on and addressed.

As data-driven prediction models advance, an increasing number of people are enjoying news personalized to their interests. The primary problem such recommendation models solve is to precisely match information with users and, in so doing, ensure that news spreads with greater efficiency. However, these techniques only help the media platform; they do not help those who produce the news. Hence, we devised a propagation framework based on a human-in-the-loop simulation that helps content authors maximize the spread of their messages through social networks. The framework works by acting on feedback provided by the simulation model. Additionally, the spread of information is formulated as a multi-objective optimization problem in which propagation is data-driven and simulated with machine learning techniques that leverage data on the historical behaviors of users. We additionally describe an implementation for this framework as an example of how the framework might be used in real life. On the practical side, the implementation uses text data from a blog to simulate the message's propagation, while, from a technical point of view, the multi-objective optimization problem is divided into an information retrieval problem and an integer programming problem, the results of which are fed back into the content editor as content operation strategies. A case study with the Sina Weibo microblog site not only validates the framework but also provides practitioners with insights into how to maximize the spread of information through social networking platforms. The results show that the proposed propagation framework is capable of increasing retweets by 7.9575 %. As an interesting aside, our experiments also show that the Weibo retweet lottery is both popular and a highly effective mechanism for increasing reposts.

Generative AI that can produce realistic text, images, and other human-like outputs is currently transforming many different industries. Yet it is not yet known how such tools might influence social science research. I argue Generative AI has the potential to improve survey research, online experiments, automated content analyses, agent-based models, and other techniques commonly used to study human behavior. In the second section of this article, I discuss the many limitations of Generative AI. I examine how bias in the data used to train these tools can negatively impact social science research--as well as a range of other challenges related to ethics, replication, environmental impact, and the proliferation of low-quality research. I conclude by arguing that social scientists can address many of these limitations by creating open-source infrastructure for research on human behavior. Such infrastructure is not only necessary to ensure broad access to high-quality research tools, I argue, but also because the progress of AI will require deeper understanding of the social forces that guide human behavior. [ABSTRACT FROM AUTHOR]