Major Social Media Platforms Research Articles

An Enhanced Twitter Corpus for the Classification of Arabic Speech Acts

Twitter has gained wide attention as a major social media platform where many topics are discussed on daily basis through millions of tweets. A tweet can be viewed as a speech act (SA), which is an utterance for presenting information, hiding indirect meaning, or carrying out an action. According to SA theory, SA can represent an assertion, a question, a recommendation, or many other things. In this paper, we tackle the problem of constructing a reference corpus of Arabic tweets for the classification of Arabic speech acts. We refer to this corpus as the Arabic Tweets Speech Act Corpus (ArTSAC). It is an enhancement of a modern standard Arabic (MSA) tweet corpus of speech acts called ArSAS. ArTSAC is more advantageous than ArSAS in terms of its richness of annotated features. The goal of ArTSAC is twofold: Firstly, to understand the purpose and intention of tweets which act in accordance with the SA theory, and hence positively influencing the development of many natural language processing (NLP) applications. Secondly, as a future goal, to be used as a benchmark annotated dataset for testing and evaluating state-of-the-art Arabic SA classification algorithms and applications. ArTSAC has been put in practice to classify Arabic tweets containing speech acts using the Support Vector Machine (SVM) classification algorithm. The results of the experiments show that the enhanced ArTSAC corpus achieved an average precision of 90.6% and an F-score of 89.6%. Substantially it outperformed the results of its predecessor ArTSAC corpus.

Prediction of Likes and Retweets Using Text Information Retrieval

Abstract Twitter is one of the major social media platforms today to study human behaviours by analysing their interactions. To ensure popularity of the tweet, the focus should be on the content of the tweet that results in numerous followings of that message with sufficient number of likes and retweets. The high quality of tweets, increases the online reputation of the users who post it. If a user can get the prediction of likes and retweets on his text before posting it on the internet, it would improve the popularity of the tweet from information sharing perspective. In this paper we employed different machine learning classifiers like SVM, Naive Bayes, Logistic Regression, Random Forest, and Neural Network, on top of two different text processing approaches used in NLP (natural language processing), namely bag-of-words (TFIDF) and word embeddings (Doc2Vec), to check how many likes and retweets can a tweet generate. The results obtained indicate that all the models performed 10-15% better with the bag-of-word technique.

Social Media and Health Care: Usage Pattern of Consumers and the Success Factors

Nowadays, healthcare organizations such as clinics and hospitals have increasingly embarked on Social Media strategies, which involve establishing an organizational presence on the major social media platforms. SM is accepted as a means to evoke patients’ knowledge on their health concerns which in turn trigger initiatives to search for healthcare products. This study is an exploration on the usage pattern of Social Media platforms for Healthcare Services by the patients or bystanders of at a tertiary level corporate health and wellness centre in Abudabi. Youth are actively searching for heath contents in social media and the information influence the significant aspects of consumer behavior. Search is more on specialty and medical terms so the content develops need to pay attention to these key aspects. Rather than health, the aspects related to wellness are getting more attention on social media plat forms. Conscious attempts while developing the content, encouraging review writing and commenting by professionals and consumers can be used to overcome the barriers regarding authenticity and quality of information.

Public Perception of Haze Weather Based on Weibo Comments.

In China, haze weather has become a major public concern and is frantically discussed by the public. Many people express their views, opinions, or complaints on social media. Effectively extracting this useful information may help to improve our understanding of how the public perceive and respond to haze, and could potentially contribute to environmental policy-making. In this paper, we study how the public perceive haze during haze weather and how this perception changes with seasons based on comment data from a major social media platform in China, Weibo, and using several social network methods, including perceptual keyword cloud mapping, complex network topology characteristics, and social perception analysis. The results showed that the public’s perception was focused on the causes of haze in spring, enjoyment of life and travel in summer, measures to prevent haze in autumn, and the adverse effects of haze on human health in winter.

Disparagement humour and gendered racism on social media in Brazil

ABSTRACT Data reveals that in 2017 in Brazil, there were 63,698 reported cases of hate speech on the internet, with one-third of these being related to racist discourses. Another study reveals that 81 per cent of the victims of racism on Facebook in Brazil are upwardly mobile black women. These discourses are often conveyed through disparagement humour, which offers the abusers the convenient excuse that their comments are simply harmless jests. This study develops a critical discourse analysis of a selection of Facebook posts, aiming to understand their embedded messages. The findings suggest, first, that racist disparagement humour is employed to challenge black women’s upward social mobility and disqualify their achievements. Secondly, they reinforce deep-seated ideologies regarding differentiated symbolic social spaces for blacks and whites in Brazil. Finally, major social media platforms represent the contemporary arena for the manifestation, dissemination, and reinforcement of racist ideologies.

The use of social media for work-related knowledge sharing by construction professionals

ABSTRACT Use of social media for work-related knowledge sharing (KS) by professionals has become widespread globally. The extent, nature and patterns of this use have been little explored to date in the construction industry. This paper aims to investigate the frequency, nature and determinants of social media use for KS among the construction professionals. The objectives are to measure the current level of use of the three major social media platforms by the construction professionals for KS; and to empirically validate an extended version of the Unified Theory of Acceptance and Use of Technology (UTAUT) that identifies the impact factors of adoption of social media for work-related KS by the construction professionals. Survey data were collected from 120 construction professionals in Australia and were analysed using Structural Equation Modelling (SEM) techniques. Findings show that Facebook and LinkedIn were used by almost half of the survey respondents. Passive KS was the most common type of KS activity. Performance Expectancy (PE), and Knowledge Sharing Self-Efficacy (KSSE) had positive impacts on the construction professionals’ intention to use (IU) social media. IU and Facilitating Conditions (FC) had positive direct impacts on the use of social media for work-related KS. The findings shed light into the potential ways of improving social media use for KS by the construction professionals.

Designing authoritarian deliberation: how social media platforms influence political talk in China

ABSTRACT Discussion is often celebrated as a critical element of public opinion and political participation. Recently, scholars have suggested that the design and features of specific online platforms shape what is politically expressed online and how. Building on these findings and drawing on 112 semi-structured qualitative interviews with information technology experts and internet users, we explain how major Chinese social media platforms differ in structure and motivation. Drawing upon a nationwide representative survey and an online experiment, we find that platforms aiming to make users a source of information through public, information-centred communication, such as the Twitter-like Weibo, are more conducive to political expression; while platforms built to optimize building social connections through private, user-centred communication, such as WhatsApp and Facebook-like WeChat, tend to inhibit political expression. These technological design effects are stronger when users believe the authoritarian state tolerates discussion, but less important when political talk is sensitive. The findings contribute to the debate on the political consequences of the internet by specifying technological and political conditions.

‘A battlefield for public opinion struggle': how does news consumption from different sources on social media influence government satisfaction in China?

ABSTRACT While the political influence of the Internet, especially social media, on people’s attitude toward their governments has been widely discussed in western democracies, the situation in authoritarian regimes such as China has not yet been adequately addressed. The current Chinese administration considers social media ‘the main battlefield for public opinion struggle’ between the official discourse and those challenging it. To provide clues about who is shaping the public opinion, this study examined how consuming news from competing information sources on social media influences Chinese citizens’ satisfaction with the central and local government. Based on a nationally representative survey of 2,882 Chinese adults, the study found that consuming news from governmental sources on the country’s major social media platforms – Weibo and WeChat – was positively associated with citizens’ satisfaction with the central government, while exposure to alternative news sources on WeChat had a negative impact on both central and local government satisfaction. Additionally, news consumption from mainstream media sources on social media did not significantly influence the public’s government satisfaction. This paper contributes to the current literature by revealing that social media do not provide a unified agenda and by emphasizing the impact of platform affordances on people’s political attitudes.

Billedindeksering og sociale medier

This article focuses on the methodologies, organization, and communication of digital image collection research that utilizes social media content. “Image” is here understood as a cultural, conventional, and commercial—stock photo—representation. Two methodologies i.e. PRISMA and Grounded Theory are employed to examine, categorize and analyze images and comprehend how humans consider them. The literature review covers research since 2005, when major social media platforms emerged. It demonstrates that the images on social media have not changed the overall direction of research into image indexing and retrieval, though new topics on crowdsourcing and tagging have emerged in it. A citation analysis includes an overview of co-citation maps that demonstrate the nexus of image research literature and the journals in which they appear. The results point to new possibilities influencing the research by providing large image collections as new testbeds for improving or testing research hypothesis on a new scale.

VESS11. Digital Footprint of Vascular Surgery Training Programs in the United States and Canada

The digital footprint of vascular residency and fellowship programs may have an impact on an applicant’s likelihood of selecting a given program. This may include content and accessibility of a particular program’s website as well as its social media presence. The goal of this study was to evaluate the online presence of all accredited vascular surgery training programs in the United States and Canada. A list of accredited vascular surgery training programs in the United States was obtained from the Accreditation Council for Graduate Medical Education and the Society for Vascular Surgery. Canadian program websites were sourced from the Canadian Society for Vascular Surgery website. Each program website was individually queried. A systematic Google search of each program was carried out to determine website accessibility. Thirty-one individual content metrics were used to appraise the websites. Three major social media platforms (Twitter, Facebook, and Instagram) were individually searched for program profiles. A total of 105 independent vascular surgery fellowship programs in the 5+2 paradigm and 55 integrated vascular surgery residency programs in the 0+5 paradigm were identified in the United States. An additional 10 Canadian programs were also identified, including 10 integrated residency programs and 4 independent fellowships. Ninety-nine percent of integrated residency and fellowship programs were accessible through Google search. Program description was also almost universally available. Significant differences between Canadian and U.S. programs were observed (Fig), including the mention of salary information (43% vs 10%; P = .039), clinic responsibilities (38% vs 90%; P = .001), teaching responsibilities (34% vs 100%; P < .0001), PD contact information (47% vs 80%; P = .045), mention of journal club (52% vs 100%; P = .003), research requirements (50% vs 90%; P = .014), and past and current research (30% vs 70% [P = .009] and 37% vs 80% [P = .008], respectively). Significant differences were found in the mention of institutions from which trainees came from (48% vs 10%; P = .021), mention of a hybrid OR (42% vs 100%; P = .0003), advertised medical student rotations (25% vs 90%; P < .0001), and social media presence (13% vs 70%; P < .0001). The overall digital footprint of the majority of training programs in the United States was small, unlike their Canadian counterparts. Although the vast majority of websites for vascular surgery training programs were accessible via simple Internet searches, they lacked information that could have been important to applicants. The significant underuse of social media platforms by American vascular surgery programs indicates a potential missed opportunity to target the millennials who make up most of the applicant pool to these programs.

Sourcing and Automation of Political News and Information During Three European Elections

Voters increasingly rely on social media for news and information about politics. But increasingly, social media has emerged as a fertile soil for deliberately produced misinformation campaigns, conspiracy, and extremist alternative media. How does the sourcing of political news and information define contemporary political communication in different countries in Europe? To understand what users are sharing in their political communication, we analyzed large volumes of political conversation over a major social media platform—in real-time and native languages during campaign periods—for three major European elections. Rather than chasing a definition of what has come to be known as “fake news,” we produce a grounded typology of what users actually shared and apply rigorous coding and content analysis to define the types of sources, compare them in context with known forms of political news and information, and contrast their circulation patterns in France, the United Kingdom, and Germany. Based on this analysis, we offer a definition of “junk news” that refers to deliberately produced misleading, deceptive, and incorrect propaganda purporting to be real news. In the first multilingual, cross-national comparison of junk news sourcing and consumption over social media, we analyze over 4 million tweets from three elections and find that (1) users across Europe shared substantial amounts of junk news in varying qualities and quantities, (2) amplifier accounts drive low to medium levels of traffic and news sharing, and (3) Europeans still share large amounts of professionally produced information from media outlets, but other traditional sources of political information including political parties and government agencies are in decline.

Political Journalists and Their Social Media Audiences: New Power Relations

Social media use is now commonplace across journalism, in spite of lingering unease about the impact the networked, real-time logic of leading social media platforms may have on the quality of journalistic coverage. As a result, distinct journalistic voices are forced to compete more directly with experts, commentators, sources, and other stakeholders within the same space. Such shifting power relations may be observed also in the interactions between political journalists and their audiences on major social media platforms. This article therefore pursues a cross-national comparison of interactions between political journalists and their audiences on Twitter in Germany and Australia, documenting how the differences in the status of Twitter in each country’s media environment manifest in activities and network interactions. In each country, we observed Twitter interactions around the national parliamentary press corps (the Bundespressekonferenz and the Federal Press Gallery), gathering all public tweets by and directed at the journalists’ accounts during 2017. We examine overall activity and engagement patterns and highlight significant differences between the two national groups; and we conduct further network analysis to examine the prevalent connections and engagement between press corps journalists themselves, and between journalists, their audiences, and other interlocutors on Twitter. New structures of information flows, of influence, and thus ultimately of power relations become evident in this analysis.

Internet regulation as media policy: Rethinking the question of digital communication platform governance

This article identifies the current global ‘techlash’ towards the major digital and social media platforms as providing the context for a renewed debate about whether these digital platform companies are effectively media companies (publishers and broadcasters of media content), and implications this has for twenty-first-century media policy. It identifies content moderation as a critical site around which such debates are being played out, and considers the challenges arising as national and regionally based regulatory options are considered for digital platforms that are ‘born global’. It considers the shifting balance between the ‘social contract’ of public interest obligations and democratic rights of free speech and freedom of expression.

Why Do People Use Social Media So Often? What are the Contributing Factors? Is Loneliness One of the Major Contributing Factor?

Nearly a decade ago, there were no major social media platforms, but during last 10 years, there has been major increase in the number of social media sites and users. On the other hand, depression, loneliness, anxiety, etc. are also becoming widespread and common among teenagers and youths of our country. According to a WHO report, 36% of the Indian population is suffering from depression, majority of them are between age 6 to 25. So, this research has been carried to find out the contributing factors among youths and teenagers for the frequent usage of mainly three social media apps: WhatsApp, Facebook and Instagram. The data was collected through a questionnaire. The sample population was 100 college going students. Through the study it was found that 65% of the participants use all the three social media apps. 29% of them used it as a tool for the expression of their thoughts, while 27% of them used it to keep in touch with their loved ones.

What do Sport and Exercise Medicine (SEM) doctors look like online? A cross-sectional exploration of the social media presence of SEM doctors in the UK

ObjectivesTo explore the social media presence of Sport and Exercise Medicine (SEM) doctors from the UK. Secondary aims were to identify whether there were any differences in years since qualification...

Comparing automated text classification methods

Online social media drive the growth of unstructured text data. Many marketing applications require structuring this data at scales non-accessible to human coding, e.g., to detect communication shifts in sentiment or other researcher-defined content categories. Several methods have been proposed to automatically classify unstructured text. This paper compares the performance of ten such approaches (five lexicon-based, five machine learning algorithms) across 41 social media datasets covering major social media platforms, various sample sizes, and languages. So far, marketing research relies predominantly on support vector machines (SVM) and Linguistic Inquiry and Word Count (LIWC). Across all tasks we study, either random forest (RF) or naive Bayes (NB) performs best in terms of correctly uncovering human intuition. In particular, RF exhibits consistently high performance for three-class sentiment, NB for small samples sizes. SVM never outperform the remaining methods. All lexicon-based approaches, LIWC in particular, perform poorly compared with machine learning. In some applications, accuracies only slightly exceed chance. Since additional considerations of text classification choice are also in favor of NB and RF, our results suggest that marketing research can benefit from considering these alternatives.

Digital Constitutionalism: Using the Rule of Law to Evaluate the Legitimacy of Governance by Platforms

Platforms govern users, and the way that platforms govern matters. In this paper, I propose that the legitimacy of governance of users by platforms should be evaluated against the values of the rule of law. In particular, I suggest that we should care deeply about the extent to which private governance is consensual; transparent; equally applied and relatively stable; fairly enforced; and respects substantive human rights. These are the core values of good governance, but are alien to the systems of contract law that currently underpin relationships between platforms and their users. Through an analysis of the contractual terms of service of 15 major social media platforms, I show how these values can be applied to evaluate governance, and how poorly platforms perform on these criteria. The values of the rule of law provide a language to name and work through contested concerns about the relationship between platforms and their users. This is a necessary precondition to an increasingly urgent task. The law of contract does not address these governance concerns, and the concept of constitutional rights does not extend to governance by private actors. Finding a way to apply these values to articulate a set of desirable restraints on the exercise of power in the digital age is the key challenge and opportunity of the project of digital constitutionalism.

Digital Constitutionalism: Using the Rule of Law to Evaluate the Legitimacy of Governance by Platforms

Platforms govern users, and the way that platforms govern matters. In this article, I propose that the legitimacy of governance of users by platforms should be evaluated against the values of the rule of law. In particular, I suggest that we should care deeply about the extent to which private governance is consensual, transparent, equally applied and relatively stable, and fairly enforced. These are the core values of good governance, but are alien to the systems of contract law that currently underpin relationships between platforms and their users. Through an analysis of the contractual Terms of Service of 14 major social media platforms, I show how these values can be applied to evaluate governance, and how poorly platforms perform on these criteria. I argue that the values of the rule of law provide a language to name and work through contested concerns about the relationship between platforms and their users. This is an increasingly urgent task. Finding a way to apply these values to articulate a set of desirable restraints on the exercise of power in the digital age is the key challenge and opportunity of the project of digital constitutionalism.

The State-of-the-Art in Twitter Sentiment Analysis

Twitter has emerged as a major social media platform and generated great interest from sentiment analysis researchers. Despite this attention, state-of-the-art Twitter sentiment analysis approaches perform relatively poorly with reported classification accuracies often below 70%, adversely impacting applications of the derived sentiment information. In this research, we investigate the unique challenges presented by Twitter sentiment analysis and review the literature to determine how the devised approaches have addressed these challenges. To assess the state-of-the-art in Twitter sentiment analysis, we conduct a benchmark evaluation of 28 top academic and commercial systems in tweet sentiment classification across five distinctive data sets. We perform an error analysis to uncover the causes of commonly occurring classification errors. To further the evaluation, we apply select systems in an event detection case study. Finally, we summarize the key trends and takeaways from the review and benchmark evaluation and provide suggestions to guide the design of the next generation of approaches.

Implicit negative link detection on online political networks via matrix tri-factorizations

ABSTRACTPolitical conversations have become a ubiquitous part of social media. When users interact and engage in discussions, there are usually two mediums available to them; textual conversations and platform-specific interactions such as like, share (Facebook) or retweet (Twitter). Major social media platforms do not facilitate users with negative interaction options. However, many political network analysis tasks rely on not only positive but also negative linkages. Thus, detecting implicit negative links is an important and a challenging task. In this work, we propose an unsupervised framework utilising positive interactions, sentiment cues, and socially balanced triads for detecting implicit negative links. We also present an online variant of it for streaming data tasks. We show the effectiveness of both frameworks with experiments on two annotated datasets of politician Twitter accounts. Our experiments show that the proposed frameworks outperform other well-known and proposed baselines. To illustrate the detected implicit negative links' contribution, we compare the community detection accuracies using unsigned and signed networks. Experimental results using detected negative links show superiority on the three datasets where the camps are known a priori. We also present qualitative evaluations of polarisation patterns between communities which are only possible in the presence of negative links.