Twitter Archive Research Articles

Social Media Archiving in Practice: A Troubled Landscape in Review

ABSTRACT The following article provides a brief overview on the current state of institutional archival practices of social media content and explores some of the key challenges the field continues to face. Social media use has developed into the leading means of human communication in the world, with billions of people utilizing multiple platforms on a daily basis to both document their individual lives and forge digital communities around various interests. Users generate exorbitant amounts of ephemeral content in the form of text, photos, videos and more across multiple platforms which are, by design, unarchivable. As a result, libraries, archives, and other cultural preservation institutions have struggled to find the proper technological tools to capture the sheer amount of data being generated, while also grappling with a variety of methodological, ethical and legal questions that the practice raises. All the while, the amount of social media data containing potentially important cultural heritage artifacts of our time continues to grow and remain at risk of being forever lost, altered or altogether erased. This article looks at several notable examples of major social media data loss that have already taken place in recent history and examines one of the largest institutional attempts to preserve social media thus far in the Library of Congress’ Twitter Archive. By exploring some of the key challenges that arose during this attempt which ultimately grounded the project, this article aims to better understand what continues to keep the practice of social media archiving at bay, and what large scale changes might be necessary to make any further progress in the field.

“I Can’t Breathe:” The Invisible Slow Violence of Breathing Politics in Minneapolis

Following the death of George Floyd in Minneapolis, his utterance, “I can’t breathe,” reverberated internationally as the world population grappled with the twin specters of life-threatening COVID-19 respiratory morbidities and mounting years under increasingly polarized racist regimes. Despite crisis fatigue, national and international outpourings of solidarity trended on social and mainstream media. However, in this moment, the legacy of structural and slow violences against the living, breathing Minneapolis–St. Paul communities of color were obscured. This article addresses transdisciplinary breathing politics in this mid-sized American city to integrate atmospheric indicators (concentrations of criteria pollutants including particulate matter and gaseous pollutants), traffic indicators (Minnesota Department of Transportation permanent traffic monitoring station data), and social indicators (community responses in newspaper and Twitter archives), ultimately making visible how Floyd’s utterance reflects much deeper patterns of stratified urban public health risks and socio-environmental airscape politics. Bullet Points of Findings Breathing politics are racialized in Minneapolis, demonstrating stark differences in traffic and air quality across neighborhoods. Through content analysis, it is shown that social media platforms like Twitter can be rich historical records for tracking local public discourse, providing valuable insight to the ways people talk about and conceive topics like environmental justice, breathing politics, and urban equity. While hashtag activism on social media flourished in 2020 to address anti-Black racism, it was neither a “tipping point” nor did it show a discernible impact on the nature of environmental justice discourse about breathing politics, despite the steep rise of #ICantBreathe. Integrating social, economic, and environmental indicators has the overarching benefit of addressing complex, lived systems.

An extensive work on comparing sentiment patterns in twitter archives between two persons

In social media starting from 2006 Twitter gets Major attention because of the text shared via Twitter causes positive or negative impacts. Hence comparing two persons twitter archives is important to analyse the similarity of words they use semantics of the reviews and finding the Trends of words. This study evaluates how the characteristics of tweets are changing over years. This study finds which words used in tweets cause a number of retweets and also returns the sentiments of tweets.

Using Machine Learning to Enhance Archival Processing of Social Media Archives

This article reports on a study using machine learning to identify incidences and shifting dynamics of hate speech in social media archives. To better cope with the archival processing need for such large-scale and fast evolving archives, we propose the Data-driven and Circulating Archival Processing (DCAP) method. As a proof-of-concept, our study focuses on an English language Twitter archive relating to COVID-19: Tweets were repeatedly scraped between February and June 2020, ingested and aggregated within the COVID-19 Hate Speech Twitter Archive (CHSTA), and analyzed for hate speech using the Generative Adversarial Network–inspired DCAP method. Outcomes suggest that it is possible to use machine learning and data analytics to surface and substantiate trends from CHSTA and similar social media archives that could provide immediately useful knowledge for crisis response, in controversial situations, or for public policy development, as well as for subsequent historical analysis. The approach shows potential for integrating multiple aspects of the archival workflow and supporting automatic iterative redescription and reappraisal activities in ways that make them more accountable and more rapidly responsive to changing societal interests and unfolding developments.

The Past Web: Exploring Web Archives

The Past Web: Exploring Web Archives

Deep learning based hashtag recommendation system for multimedia data

This work aims to provide a novel hybrid architecture to suggest appropriate hashtags to a collection of orpheline tweets. The methodology starts with defining the collection of batches used in the convolutional neural network. This methodology is based on frequent pattern extraction methods. The hashtags of the tweets are then learned using the convolution neural network that was applied to the collection of batches of tweets. In addition, a pruning approach should ensure that the learning process proceeds properly by reducing the number of common patterns. Besides, the evolutionary algorithm is involved to extract the optimal parameters of the deep learning model used in the learning process. This is achieved by using a genetic algorithm that learns the hyper-parameters of the deep architecture. The effectiveness of our methodology has been demonstrated in a series of detailed experiments on a set of Twitter archives. From the results of the experiments, it is clear that the proposed method is superior to the baseline methods in terms of efficiency.

An autoethnographic account of the use of Twitter for professional development by novice academic

This paper presents an exploratory autoethnographic study of my use of Twitter to support my professional development. It investigated three research questions that focus on my approach to using Twitter, experiences and challenges, and management strategies. Using reflective narrative and data from my Twitter archive, five themes were developed using thematic, lexical, and social network analysis. These themes are: 1) constructing a personal learning network, 2) managing an evolving space, 3) serendipitous learning and staying in touch, 4) spontaneous engagement and opportunistic collaboration and 5) adopting helpful behaviours. These results suggest my use of Twitter evolved over time into a personal and networked space that has positively impacted my development as a novice academic. This networked space offered continuous opportunities for learning, collaboration and engagement. Though several challenges were experienced, deliberate strategies mitigated these challenges. While this study found similarities between my experiences and those of others from the literature, it offers potentially unique strategic insights into the use of Twitter for effective professional development.Keywords: Twitter; social networks; autoethnography; professional development; higher education; novice academicsPart of the special issue Autoethnography in online doctoral education <https://doi.org/10.21428/8c225f6e.9415e58d>

The breathing human infrastructure: Integrating air quality, traffic, and social media indicators

Outdoor air pollution is a complex system that is responsible for the deaths of millions of people annually, yet the integration of interdisciplinary data necessary to assess air quality's multiple metrics is still lacking. This case study integrates atmospheric indicators (concentrations of criteria pollutants including particulate matter and gaseous pollutants), traffic indicators (permanent traffic monitoring station data), and social indicators (community responses in Twitter archives) representing the interplay of the three critical pillars of the United Nations' Triple Bottom Line: environment, economy, and society. During the watershed moment of the COVID-19 pandemic lockdowns in Florida, urban centers demonstrated the gaps and opportunities for understanding the relationships, through correlations rather than causations, between urban air quality, traffic emissions, and public perceptions. The relationship between the perception and the traffic variables were strongly correlated, however no correlation was observed between the perception and actual air quality indicators, except for NO2. These observations might consequently infer that traffic serves as people's proxy for air quality, regardless of actual air quality, suggesting that social media messaging around asthma may be a way to monitor traffic patterns in areas where no infrastructure currently exists or is prohibited to build. It also indicates that people are less likely to be reliable sensors to accurately measure air quality due to bias in their observations of traffic volume and/or confirmation biases in broader social discourse. Results presented herein are of significance in demonstrating the capacity for interdisciplinary studies to consider the predictive capacities of social media and air pollution, its use as both lever and indicator of public support for air quality legislation and clean-air transitions, and its ability to overcome limitations of surface monitoring stations.

Trump's COVID-19 tweets and Dr. Fauci's emails.

The paper features an analysis of former President Trump’s early tweets on COVID-19 in the context of Dr. Fauci’s recently revealed email trove. The tweets are analysed using various data mining techniques, including sentiment analysis. These techniques facilitate exploration of content and sentiments within the texts, and their potential implications for the national and international reaction to COVID-19. The data set or corpus includes 159 tweets on COVID-19 that are sourced from the Trump Twitter Archive, running from 24 January 2020 to 2 April 2020. In addition we use Zipf and Mandelbrot’s power law to calibrate the extent to which they differ from normal language patterns. A context for the emails is provided by the recently revealed email trove of Dr. Fauci, obtained by Buzzfeed on 1 June 2021 obtained under the Freedom of Information Act.

Ideological Manipulation in Twitter Communication: A Critical Stylistic Analysis of Donald Trump's Tweets

Donald Trump’s use of Twitter as a modality to defame opponents, antagonize media outlets and even glorify violence is an enduring legacy for political campaigns, presidential rhetoric and argumentative debates. This nontraditional use of social media as a political communication tool has invited Twitter’s fact-checking editorial decisions, alienated some of Trump’s supporters and attracted worldwide criticism. Using purposive sampling, the present paper employs the ten textual-conceptual functions of critical stylistics to analyze a dataset of Trump’s tweets on domestic and international political issues published between 2011 and 2020 and assembled from the monitor corpus Trump Twitter Archive. The critical stylistic analysis aims at uncovering Trump’s ideological outlook by identifying the extra layer of meaning in which the ideological evaluation is structured and exposing the way in which the resources of language are strategically deployed to influence and ideologically manipulate Trump’s followers’ experience of reality. Analysis reveals a network of lexical, syntactic, semantic and pragmatic choices underlying Trump’s seemingly simple rhetoric. It signposts his ideological evaluation and constructs a world for his followers to desire, believe or fear. The study extends the application of critical stylistics to microblogging channels, with implications both for the linguistic make-up of political communication in Web 2.0 contexts and for the explanatory power of critical stylistics..

Twitter as research data Tools, costs, skill sets, and lessons learned.

Scholars increasingly use Twitter data to study the life sciences and politics. However, Twitter data collection tools often pose challenges for scholars who are unfamiliar with their operation. Equally important, although many tools indicate that they offer representative samples of the full Twitter archive, little is known about whether the samples are indeed representative of the targeted population of tweets. This article evaluates such tools in terms of costs, training, and data quality as a means to introduce Twitter data as a research tool. Further, using an analysis of COVID-19 and moral foundations theory as an example, we compared the distributions of moral discussions from two commonly used tools for accessing Twitter data (Twitter's standard APIs and third-party access) to the ground truth, the Twitter full archive. Our results highlight the importance of assessing the comparability of data sources to improve confidence in findings based on Twitter data. We also review the major new features of Twitter's API version 2.

Quality assessment of crowdsourced social media data for urban flood management

Urban flooding can cause widespread devastation in terms of loss of life and damage to property. As such, monitoring urban flood evolution is crucial in identifying the most affected areas, where emergency response resources should be directed. Flood monitoring through airborne or satellite remote sensing is often limited due to weather conditions and urban topography. In contrast, crowdsourced data is not affected by weather or topography, and they hence offer great potential for urban flood monitoring through real-time information shared by individuals. Despite the benefits, there is no guarantee of quality associated with crowdsourced data, which hampers its usability. In this paper, we present and evaluate two different approaches (binary logistic regression and fuzzy logic) to assess the quality of crowdsourced social media data retrieved from the public Twitter archive. Input variables were constructed based on Twitter metadata and spatiotemporal analysis. Both models were trained and tested using actual flood-related information Tweeted during three consecutive years of flooding in Phetchaburi City, Thailand (2016 to 2018), and produced good results. The fuzzy logic approach is shown to perform better, however its implementation involves significantly more subjectivity. The ability to assess data quality enables the uncertainty associated with crowdsourced social media data to be estimated, which allows this type of data to supplement conventional observations, and hence improve flood management activities.

Tweeted, deleted: An exploratory study of the US government’s digital memory holes

Deletion is part of the Internet’s history and predates Twitter. Today, research on the laws underlying and facilitating government social media use and deletion practices has remained limited. The question of how government agencies create their own Twitter archives and subsequently institutionalize cultural memory has also been largely unexplored. Drawing on a US case study, I argue that the lack of a standardized federal policy has led to the creation of myriad digital “memory holes” of varying porosities. I show that, when systematically drafted and deployed, research based on the US Freedom of Information Act (FOIA) can serve as a generative method of unearthing deleted tweets and memory narratives that might otherwise be inaccessible. Finally, I build on the US case study by offering pathways for other new media scholars to examine and trace tweeting and deleting by government employees in other jurisdictions.

#MeToo and the Politics of Collective Healing: Emotional Connection as Contestation

Abstract Participants in the #MeToo movement on Twitter expressed emotions like rage, pain, and solidarity in their personal accounts of sexual violence. This article explores the digital circulation of these affects and considers how the outpouring of tweets about sexual harassment and abuse contribute to a feminist politics centered on collective healing. The particular emotions expressed in the #MeToo Twitter archive subvert the logics of quantification and visibility that undergird popular feminism and the attention economy, and produce an affective excess that works toward movement founder Tarana Burke's original project of “mass healing.” At a moment wherein popular feminism emphasizes individual empowerment and consumption, and carceral feminism relies on criminalization and incarceration, the #MeToo movement's focus on shared emotions represents the potential for a feminist politics rooted in collective support and restorative justice.

E-Memory Choice Architecture

Twitter archiving systems have been developed to preserve users' tweets. The available methods of organizing tweets for curation include the hashtag, user ID, and keywords. These can be viewed as memory encoding symbols supporting future retrieval of users' social media memories. As Twitter has become a global social media platform, online Twitter archiving systems have transformed from an open platform for archiving tweets to an integrated service managing multiple accounts across platforms. With the changing business models of Twitter archiving systems, usage data has become unavailable publicly. This study collected historical usage data from the API of an online Twitter archiving system, TwapperKeeper, before its acquisition by Hootsuite in September 2011. The valuable system usage data allowed this study to examine the tweet archiving preferences of early Twitter adopters. By mapping adoption-diffusion and use-diffusion models into the web information architecture of the online archiving system, this study analyzed user choice architecture through the system function use.

Tracking the history and evolution of entities: entity-centric temporal analysis of large social media archives

How did the popularity of the Greek Prime Minister evolve in 2015? How did the predominant sentiment about him vary during that period? Were there any controversial sub-periods? What other entities were related to him during these periods? To answer these questions, one needs to analyze archived documents and data about the query entities, such as old news articles or social media archives. In particular, user-generated content posted in social networks, like Twitter and Facebook, can be seen as a comprehensive documentation of our society, and thus meaningful analysis methods over such archived data are of immense value for sociologists, historians and other interested parties who want to study the history and evolution of entities and events. To this end, in this paper we propose an entity-centric approach to analyze social media archives and we define measures that allow studying how entities were reflected in social media in different time periods and under different aspects, like popularity, attitude, controversiality, and connectedness with other entities. A case study using a large Twitter archive of four years illustrates the insights that can be gained by such an entity-centric and multi-aspect analysis.

Archiving and Preserving Social Media at the Library of Congress: Institutional and Cultural Challenges to Build a Twitter Archive

AbstractInstitutional archiving of media is neither new nor strange. The United States Library of Congress has been preserving printed materials, newspapers, photographs, film, and even websites for decades—if not centuries. After seven years, in later 2017, the initiative to build a Twitter Archive came to a halt. Through a textual analysis of policy papers, preservation theories and press releases, this study illustrates the social, cultural, and symbolic challenges of institutional archiving of digital media.

O9.2. IDENTIFYING PSYCHOTIC SYMPTOMS AND PREDICTING RELAPSE THROUGH SOCIAL MEDIA

BackgroundThe internet and social media provide an unprecedented opportunity to transform early psychosis intervention services. This study aimed to capture concerning patterns of social media activity associated with the onset and persistence of psychotic symptoms.MethodsFacebook and Twitter archives were extracted from over 150 participants with psychotic disorders, mood disorders and healthy controls. Machine learning was used to build classifiers aiming to identify patterns and distinguish between groups.ResultsLinguistic analysis of Twitter commentary identified significantly increased use of interpersonal pronouns (p < 0.001), decreased emphasis on friendship (p < 0.001) and increased emphasis on health (p < 0.001) in individuals with psychosis. Preliminary classifiers correctly recognized participants with psychotic disorders (n=62) from healthy controls (n=24) with an average accuracy of 80% and distinguished participants with psychosis from those with mood disorders (n=39) with an average accuracy of 70%. Further analysis identified shifts in language use of participants with psychosis who experience a relapse (n=18) including significant increases in the use of swearing (p<0.05), first-person pronouns (p<0.05) and negations (p<0.05). We additionally identified significant differences in the profile pictures (p<0.005) and structure of messages posted (p<0.005) by youth with psychosis who experienced a psychotic relapse.DiscussionIdentifying markers in social media activity associated with worsening psychotic symptoms offers the prospect that social media may be a clinically useful tool to identify patients in the earliest phases of relapse.

Evaluating the Network: A Workflow for Tracking Twitter Interactions Using Social Networking Analysis

Networking plays an important role in research projects to build a community and audience around a research area. Using social media is popular in project communication as it provides the ability to engage with a group of followers daily. Such online networking tools provide the advantage of providing near-realtime data, which can be used to evaluate communication and networking success. This does, however, present a problem of what data can be collected in a reasonable timeframe to usefully feed back into the project’s strategies. This paper presents the evaluation approach the Research on Open Educational Resources for Development in the Global South (ROER4D) project has applied to understanding the project’s reach and engagement on the social networking site Twitter. This evaluation considered: 1) To what extent has ROER4D built a network of OER scholars? and 2) Which of ROER4D’s actions have been most effective at growing the network? Using a Utilization Focused Evaluation framework, this evaluation established a workflow through which the project’s Twitter network growth and snapshots of key online interactions could be tracked over time. Systematic measuring systems were set up to collect, select and analyse the most useful data to inform the project from three sources: Twitter analytics, Twitter Archiving Google Spreadsheet, and NodeXL. The social networking analysis shows the extent to which the project is building a network of scholars and what may have contributed to this growth. The feasibility of this networking evaluation approach for use by other projects, especially in the Global South, as well as the considerations when analysing Twitter data and metrics from accounts that would generally appeal to, in this case, the open education and OER communities, are highlighted.

Web 2.0 user knowledge and the limits of individual and collective power

This paper revisits and builds upon a prominent theme in First Monday’s 2008 “Critical perspectives on Web 2.0” special issue: exploration and critique of the claim that Web 2.0 “... provides novel opportunities for the articulation of individual and collective social power by enhancing participation in media production and cultural expression” (Zimmer, 2008). This article refreshes the critique by examining how impediments to users’ technical knowledge about how Web 2.0 platforms function affect user power.After reviewing how the scholarly literature on Web 2.0 accounts for the user knowledge/user power connection, this paper suggests Braman’s (2006) concept of “informational power” as a useful heuristic for exploring how access to information about how Web 2.0 platforms function operates as a “genetic” facet of user power. Then, using Braman’s framework, the article reviews two cases of struggle over informational power in regards to the Twitter platform: controversy over the Library of Congress Twitter archive and Occupy Wall Street protesters inaccurately accusing Twitter of censorship because of a misunderstanding about “Trending Topics.” Through the application of Braman’s framework to these cases, it becomes clear how users’ prospects for developing technical knowledge about the platforms shape the limits and horizons of individual and collective power.