Major Social Media Platforms Research Articles

Large Language Model in Creative Work: The Role of Collaboration Modality and User Expertise

Since the launch of ChatGPT in December 2022, large language models (LLMs) have been rapidly adopted by businesses to assist users in a wide range of open-ended tasks, including creative work. Although the versatility of LLM has unlocked new ways of human-artificial intelligence collaboration, it remains uncertain how LLMs should be used to enhance business outcomes. To examine the effects of human-LLM collaboration on business outcomes, we conducted an experiment where we tasked expert and nonexpert users to write an ad copy with and without the assistance of LLMs. Here, we investigate and compare two ways of working with LLMs: (1) using LLMs as “ghostwriters,” which assume the main role of the content generation task, and (2) using LLMs as “sounding boards” to provide feedback on human-created content. We measure the quality of the ads using the number of clicks generated by the created ads on major social media platforms. Our results show that different collaboration modalities can result in very different outcomes for different user types. Using LLMs as sounding boards enhances the quality of the resultant ad copies for nonexperts. However, using LLMs as ghostwriters did not provide significant benefits and is, in fact, detrimental to expert users. We rely on textual analyses to understand the mechanisms, and we learned that using LLMs as ghostwriters produces an anchoring effect, which leads to lower-quality ads. On the other hand, using LLMs as sounding boards helped nonexperts achieve ad content with low semantic divergence to content produced by experts, thereby closing the gap between the two types of users. This paper was accepted by D. J. Wu, information systems. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2023.03014 .

Impact of Social Media Usage on Sleep and Eye Health: A Survey Study Among University Students

Social media has become essential for university students, particularly during the pandemic, however, excessive late-night use can disrupt sleep due to screen light and impacting overall well-being and academic performance. Therefore, it is important to examine the purposes for which social networking sites are used and the effect of social media on sleep health among students. For the present study exploratory research design was adopted and the data was collected through google forms. The results indicated that, majority of the students were female, highest per cent belonged to 20-24 years age group, nearly seventy per cent were under graduates, students were using all major social media platforms daily to know the latest news and to chat with friends. The high social media usage of social media had high impact on their vison and led to late bedtimes.

Visualizing Mental Health Insights: A Pipeline from Social Media to Chernoff Faces.

This study proposes an approach for analyzing mental health through publicly available social media data, employing Large Language Models (LLMs) and visualization techniques to transform textual data into Chernoff Faces. The analysis began with a dataset comprising 15,744 posts sourced from major social media platforms, which was refined down to 2,621 posts through meticulous data cleaning, feature extraction, and visualization processes. Our methodology includes stages of Data Preparation, Feature Extraction, Chernoff Face Visualization, and Clinical Validation. Dimensionality reduction techniques such as PCA, t-SNE, and UMAP were employed to transform complex mental health data into comprehensible visual representations. Validation involved a survey among 60 volunteer psychiatrists, underscoring the visualizations' potential for enhancing clinical assessments. This work sets the stage for future evaluations, specifically focusing on a combined features method to further refine the visual representation of mental health conditions and to augment the diagnostic tools available to mental health professionals.

Understanding news-related user comments and their effects: a systematic review

There has been growing interest in research on news-related user comments. Here we conduct the first systematic review of this literature—quantitatively and qualitatively (248 studies)—that covers the entire communication process (content analyses, surveys, experiments). Results indicate a focus on online news articles (vs videos) and little consideration for major social media platforms (Instagram, TikTok). Research often assesses incivility in comments but offers conflicting conclusions on the actual level of incivility in comment threads—and seldom considers how to effectively combat any incivility. We propose four priorities for future work: more comparative and longitudinal approaches; exploring social media and video content; examining platform design, content moderation and artificial intelligence; and implementing measures to reduce incivility and protect the integrity of journalism.

Social Media in Teaching English for EFL Students: A Review of Challenges and Suggestions

This research paper examines the use of social media in teaching English as a foreign language (EFL) to students. The study aims to identify the major social media platforms utilized for EFL teaching and learning, assess their effectiveness, determine the challenges faced by students and teachers, and pinpoint best practices for integrating social media into English language instruction. The paper provides a review of the literature on the role of social media in English language learning and teaching. It highlights how platforms like Facebook, Twitter, Instagram, and YouTube can enhance various English language skills, including listening, speaking, reading, and writing. The study also discusses the advantages of using social media for language learning, such as providing opportunities for authentic communication, facilitating collaborative learning, and increasing learner motivation and engagement. Furthermore, the research paper explores the challenges associated with integrating social media into EFL teaching, such as technical issues, pedagogical concerns, and student/teacher resistance. It then suggests best practices for effectively leveraging social media in the EFL classroom, including selecting appropriate platforms, designing engaging activities, and providing necessary training and support for both teachers and students. The findings of this study can offer valuable insights for EFL teachers and institutions on how to effectively incorporate social media into their English language teaching practices, ultimately enhancing the learning experience and outcomes for EFL students.

Measuring the relationship between museum attributes and visitors: An application of topic model on museum online reviews.

In recent years, cultural tourism has increasingly embraced museum visits. Museums serve as both cultural heritage guardians and integral parts of tourist destinations, significantly impacting visitor satisfaction and experience. Moreover, online museum reviews have become a crucial indicator of museum service quality, visitor experience, and public feedback in the digital age. An analysis of online reviews on major tourism websites and social media platforms can assist museums in developing appropriate management strategies. This study employed the structural topic model (STM) to analyze online museum reviews, identifying three primary attributes of museums and visitors' personal experiences, as well as 19 sub-attributes. The study confirmed that core offerings have a positive impact on visitor experience and satisfaction, while peripheral services and overall ambiance are also positively related to visitor experience and satisfaction. Furthermore, the results of structural equation modeling demonstrated that visitors' personal experiences have a positive impact on satisfaction. The results of structural equation modeling analysis support all seven hypothesized relationships. These findings will assist museum managers in developing effective management strategies and future plans.

The Social Media as Means for Self-Expression Against Social Injustice and Agigitation in Nigeria

This work strives to know the place of social media in social assertiveness. The research method is survey. The study uses convenience sampling to select 50 undergraduate students from Adekunle Ajasin University, Akungba-Akoko. The findings of the study indicate that the majority of the respondents adopt Facebook and WhatsApp as major social media platforms for expressing their disgust and grievances against social ills and unfavorable government policies. It is also discovered from the study that the majority of the respondents do not expect agents of the government to care about their voices but still find a bit of relief from expressing their grievances through social media. The study concludes that, in spite of this position, the social media remain an escape avenue and means of social assertiveness for the governed.

Extracting Retrospective Information through the Public Sharing of Female Golfers' Contributions in a Digital Environment: An Analytical Perspective of Their Impact on Shaping Golf Communities

Abstract Social media are an integrated part of everyday life. They are creating a new paradigm of information consumption. The era of popularity of searching for information in large texts is irreversibly gone. The current model of communication and information distribution favours visual content. One of the major social media platforms that specializes in expressive visual content is Instagram. The essence of a communication idea is effectively rendered in a unique image. Thus, it becomes a key tool for the transmission of communication. In the above technical paper, we present empirical data that investigate the impact of Instagram posts of three professional women golfers during the calendar year 2022. The first one is the most followed golfer on social networks, the second one ranks among the current best professional golfers and the third one holds the position of the president of the Slovak Ladies Golf Association. These prominent figures in golf create a significant impact on the social opinion of the sport. The central aim of the present study is to comprehensively understand how certain types of visual content, also created through AI, elicit maximum audience engagement on social media. We present the results in the context of correlating the influence of different factors on the level of viewer interaction. The value is expressed through the number of likes awarded and the size of the membership base of the affected accounts.

HIV self-testing adoption and post-test linkage to care among men who have sex with men in China: a nationwide cross-sectional survey.

HIV self-testing (HIVST) was recommended to improve HIV testing services. China initiated some of the first HIVST pilots in the world, providing a unique opportunity for implementation research. We aim to investigate HIVST adoption and its following linkage to care among Chinese men who have sex with men (MSM). Data were collected using an online questionnaire distributed on major social media platforms in 2018, one year after HIVST was officially endorsed and allowed for sale. MSM who were at least 16 years old, assigned as male at birth, and ever tested for HIV were eligible. Primary outcome, adoption was defined as ever use of HIVST. Bivariate and multivariable logistic regressions were performed to explore the association between HIVST adoption and sociodemographic and behavioral factors. Linkage to care was also described via the following sequential events as indicators: (1) receiving result after recent test (2), seeking care from healthcare facility if test result was positive or indeterminate, and (3) delayed time before seeking care. A total of 540 participants were included with an average age of 27.4 ± 6.6. Most were never married (87.4%) and half completed college (52.2%). Overall, 75.2% had adopted HIVST. Self-test kits were commonly obtained from community-based organizations (54.4%) and from online (46.6%). HIVST adoption was positively associated with having college or higher education (OR = 1.66, 95%CI: 1.07-2.57), and negatively associated with age older than 30 (AOR = 0.52, 95%CI: 0.32-0.84). Adoption was not associated with other socio-demographic or behavioral factors. After receiving HIV-positive or indeterminate results, 25/25 of HIVST adopters sought care while 3 out of 7 (42.9%) non-adopters sought care (p < 0.001). Delays before seeking care were not significantly different between HIVST adopters compared to non-adopters (P = 0.366). Many MSM adopted HIVST shortly after its launch. Our findings suggested that HIVST linkage to care is promising among MSM in China. Integration of HIVST with other essential sexual health services is needed.

Government Health Communication During the COVID-19 Pandemic: A BERT Topic Modeling Approach

Different levels of government agencies have exerted great effort to communicate with the public during the Covid-19 pandemic on multiple social media platforms. This study uses BERT topic modeling, an artificial intelligence model, to extract topics from various public health agencies of cities, states and the federal government on Twitter and Facebook for the years 2020 and 2021. We contrast and compare major topics addressed by these agencies related to Covid-19 and the pandemic across the two major social media platforms. The findings show how we can employ BERT topic modeling to extract social media topics during a health emergency and evaluate the extent to which topics covered by these agencies address the major social and health concerns of the pandemic.

Navigating influence: Unraveling the impact of micro-influencer attributes on consumer choices in the Chinese social media

This study aims to explore the relationship between consumer purchasing behavior and key micro-influencer attributes, including knowledge, entertainment value, credibility, and transparency, within the context of Chinese social media platforms. The paper adopts a quantitative approach, employing partial least squares structural equation modeling (PLS-SEM) to analyze the intricate relationships among latent variables. The respondents comprise active users of major Chinese social media platforms, such as Weibo and Xiaohongshu. For primary data collection, 329 respondents were surveyed online, utilizing a convenient sampling method as part of non-probability sampling. Data collection spanned four weeks, and participants were given the option to respond in either English or Mandarin. The findings suggest significant associations between consumer purchasing behavior and micro-influencer attributes. Specifically, knowledge, entertainment value, credibility, and transparency exhibit varying degrees of influence on consumer behavior within the Chinese social media landscape. The p-value for H1, H2, H3, and H7 appeared as 0.000 and shows that these are the highly significant relations, whereas the p-value for H3 (0.019), for H5 (0.001), and for H6 (0.028) shows that these relations play a moderate role in the proposed model. Elucidating the role of key attributes provides valuable insights for marketers and businesses seeking to leverage micro-influencer marketing strategies effectively in this rapidly evolving digital landscape.

Analyzing Burn-Related Content on Social Media Platforms: A Study of Creator Types, Content Focus, and Engagement Metrics.

Social media offers a readily available, cost-effective way for medical experts to disseminate knowledge and shape public health outcomes but also allows for the spread of misinformation. This study aims to analyze burn-related material on social media by creator, content type, and engagement. Facebook, TikTok, and X (formerly Twitter) were queried with the following search terms: "burn," "burn injury," "burn recovery," and "burn treatment." Identified accounts were then manually screened for relevance. Year of creation and engagement metrics were collected. Accounts were categorized by content and creator type. Data was reported using descriptive statistics and visualized graphically to explore trends. Our search yielded 434 profiles, 234 of which met inclusion criteria. TikTok had the most engagement at a median of 43,500 followers per account, with 38.3% of accounts focusing on individual experiences of burn survivors primarily on personal accounts (48.3%). In contrast, content on Facebook was related to promotion of medical services (36.9%), where the most represented creator type was medical centers (33.6%). Nonprofits made up 40.4% of accounts on Twitter/X and more than a third of the content focused on patient advocacy, support, or burn prevention (36.5%). Important topics like burn education, prevention, and social support are lacking on major social media platforms. Engagement from burn care organizations and burn experts on social media is necessary. The findings of this study may guide advocates in the burn community on where and how to disseminate information in social media.

Clip-GCN: an adaptive detection model for multimodal emergent fake news domains

Emergent news is characterized by few labels, and news detection methods that rely on a large number of labels are difficult to apply to learned features for emerging events and are ineffective in coping with less labeled emergent news detection. To address the challenge of limited labeled data, this study first establishes a scenario for detecting breaking news, ensuring that the domain of detecting events is distinct from the domain of historical events. Secondly, we propose the Clip-GCN multimodal fake news detection model. The model utilizes the Clip pre-training model to perform joint semantic feature extraction of image-text information, with text information as the supervisory signal, which solves the problem of semantic interaction between modalities. Meanwhile, considering the domain attributes of news, the model is trained to extract inter-domain invariant features through Adversarial Neural Network ideation, and intra-domain knowledge information is utilized through graph convolutional networks (GCN) to detect emergent news. Through an extensive number of experiments on Chinese and English datasets from two major social media platforms, Weibo and Twitter, it is demonstrated that the model proposed in this paper can accurately screen multimodal emergent news on social media with an average accuracy of 88.7%. The contribution of this study lies not only in the improvement of model performance but also in the proposal of a solution for the challenges posed by limited labels and multimodal breaking news. This provides robust support for research in related fields.

Fertility education on Instagram: advertisements vs. educational content analysis for posts in Bahasa (Indonesian language).

As one of Indonesia's major social media platforms, Instagram provides abundant information, including fertility topics. However, fertility information in Bahasa (Indonesian language) had never been mapped before. This study aims to determine the prevalence, authorship, and types of fertility-related information in Bahasa shared on Instagram using hashtag and content analysis. Additional outcomes included a comparison of post content stratified by author type (healthcare providers vs. non-healthcare providers). Five hashtags comprising fertility terms in Bahasa (Indonesian language) were derived. Content analysis was performed on the top 100 posts for each hashtag to determine the authorship and content type. The contents posted by healthcare providers were analyzed and compared to those published by non-healthcare providers. Our search yielded 3883000 posts for the five most popular hashtags. The three most popular hashtags were 'promil', 'program hamil' and 'bayitabung'. Authorship of the top posts for each hashtag was predominantly by for-profit healthcare institutions (27.6%), followed by commercial alternative/herbal medicine (20.2%), patients (20%), physicians (11.8%), allied health professionals (3.2%), professional societies (0.8%), and others (16.4%). Of these posts, 35.4% were advertisements, 28% were related to patient experience, and 18% were educational. Healthcare provider groups were more likely to author educational posts than advertisements, while non-healthcare providers were more likely to post content for commercial purposes (P < 0.001). Fertility-related posts were dominated by for-profit healthcare institutions and alternative/herbal medicines for commercial purposes. The physician group authored more of the educational content. There is plenty of room for optimal utilization of social platforms as educational media.

Evaluating public opinions: informing public health policy adaptations in China amid the COVID-19 pandemic

Public concern regarding safety policies serious consequences is anticipated to persist over an extended duration. A study examining a case of rapid public health policy adaptation in China during the COVID-19 epidemic was conducted by gathering public opinion data from major social media platforms. A systematic approach to comprehend public opinion was developed. Five fundamental elements and four dimensions were delineated. An indicator system was established utilizing the K-means text clustering model. Public prediction, expectation, and their evolution underlying public concern were elucidated employing TF–IDF text mining models. The HMM elucidated the way public opinion influences policy adjustments. The findings underscore that public concern regarding enduring events undergoes temporal shifts, mirroring the evolution of public opinion towards policy. Public opinion aroused by both the original event and derived events collaboratively influence policy adjustments. In China, public opinion serves as a mechanism for policy feedback and oversight; notably, negative public sentiment plays a pivotal role in expediting policy transitions. These findings aid in refining policies to mitigate emergencies through a feedback loop, thereby averting the emergence of safety risks such as social unrest prompted by public opinion.

A COMPARATIVE REVIEW OF AI-GENERATED IMAGE DETECTION ACROSS SOCIAL MEDIA PLATFORMS

The proliferation of images generated by artificial intelligence (AI) has significantly impacted the digital landscape, especially on social media platforms where the distinction between natural and synthetic content is increasingly blurred. This study embarks on a comparative review of the strategies used by major social media platforms—Facebook/Instagram, Twitter, TikTok, and YouTube—to detect AI-generated images. Employing a comprehensive methodology that includes a systematic review of academic literature, analysis of platform policies, and expert interviews, this research assesses the effectiveness of various detection methods, ranging from sophisticated AI tools to user reporting mechanisms. The findings reveal diverse approaches: Facebook and Instagram utilise a blend of AI detection and human moderation; Twitter integrates machine learning algorithms with user reports; TikTok emphasises AI tools within moderation workflows and educational initiatives; and YouTube relies on its Content ID system alongside AI analysis. The study highlights the critical role of effective detection systems in maintaining content authenticity and user trust, underscoring the importance of balancing automated detection with human oversight. The ongoing development and refinement of these technologies, alongside collaborative efforts and evolving regulatory frameworks, are identified as essential for ensuring a trustworthy digital environment. This research contributes to the discourse on digital integrity, offering insights into the complexities of safeguarding social media ecosystems against the challenges posed by AI-generated content.

The Gift of Gab: A Netnographic Examination of the Community Building Mechanisms in Far-Right Online Space

ABSTRACT Major social media platforms have recently taken a more proactive stand against harmful far-right content and pandemic-related disinformation on their sites. However, these actions have catalysed the growth of fringe online social networks for participants seeking right-wing content, safe havens, and unhindered communication channels. To better understand these isolated systems of online activity and their success, the study on Gab Social examines the mechanisms used by the far right to form an alternative collective on fringe social media. My analysis showcases how these online communities are built by perpetuating meso-level identity-building narratives. By examining Gab’s emphasis on creating its lasting community base, the work offers an experiential examination of the different communication devices and multimedia within the platform through a netnographic and qualitative content analysis lens. The emergent findings and discussion detail the far right’s virtual community-building model, revolving around its sense of in-group superiority and the self-reinforcing mechanisms of collective. Not only does this have implications for understanding Gab’s communicative dynamics as an essential socialisation space and promoter of a unique meso-level character, but it also reflects the need for researchers to (re)emphasise identity, community, and collectives in far-right fringe spaces.

The role of a major social media platform on students’ academic performance: Perception versus reality

The social media landscape is constantly evolving; new platforms emerge, and existing platforms change their functionality. While a robust body of literature exists on the influences of social media on students’ academic outcomes, most studies have failed to differentiate between specific social media platforms. Further, most research in this field focuses on how one variable (e.g., time spent on social media per day) relates to students’ GPA, giving an incomplete picture of how social media relate to student outcomes. The current study aimed to (a) investigate the intricate relationship between social media usage, time spent on schoolwork, and academic performance in college students; (b) confirm the rise of TikTok use among college students; and (c) understand college students’ perceptions of how their major social media platform influences their academic performance. Data were collected from a sample of undergraduate students in the USA (n=306). While the time spent on social media was negatively correlated with GPA (r=-.16, p&amp;lt;.001); time spent on schoolwork had no effect on GPA (r=.03, p=.580). Further, the time allocated to social media usage positively correlated with the time devoted to schoolwork (r=.14, p=.020), suggesting the pervasiveness of social media multitasking among college students. TikTok was the most commonly used social media platform, particularly by women. However, while TikTok-favoring students were more likely to think their GPA would be higher were they off social media, their GPA was not significantly different from other users who favored Snapchat and Instagram as their primary social media platforms, implying a discrepancy between student perception and reality. This may be because the TikTok videos are very short, thus one may watch a high number of them in a row and assume they have spent a long time on the app, which may have not been the case.

Multidimensional sentiment analysis method on social media data: comparison of emotions during and after the COVID-19 pandemic

PurposeThis study aims to elucidate the dynamics of social media discourse during global health events, specifically investigating how users across different platforms perceive, react to and engage with information concerning such crises.Design/methodology/approachA mixed-method approach was employed, combining both quantitative and qualitative data collection. Initially, thematic analysis was applied to a data set of social media posts across four major platforms over a 12-month period. This was followed by sentiment analysis to discern the predominant emotions embedded within these communications. Statistical tools were used to validate findings, ensuring robustness in the results.FindingsThe results showcased discernible thematic and emotional disparities across platforms. While some platforms leaned toward factual information dissemination, others were rife with user sentiments, anecdotes and personal experiences. Overall, a global sense of concern was evident, but the ways in which this concern manifested varied significantly between platforms.Research limitations/implicationsThe primary limitation is the potential non-representativeness of the sample, as only four major social media platforms were considered. Future studies might expand the scope to include emerging platforms or non-English language platforms. Additionally, the rapidly evolving nature of social media discourse implies that findings might be time-bound, necessitating periodic follow-up studies.Practical implicationsUnderstanding the nature of discourse on various platforms can guide health organizations, policymakers and communicators in tailoring their messages. Recognizing where factual information is required, versus where sentiment and personal stories resonate, can enhance the efficacy of public health communication strategies.Social implicationsThe study underscores the societal reliance on social media for information during crises. Recognizing the different ways in which communities engage with, and are influenced by, platform-specific discourse can help in fostering a more informed and empathetic society, better equipped to handle global challenges.Originality/valueThis research is among the first to offer a comprehensive, cross-platform analysis of social media discourse during a global health event. By comparing user engagement across platforms, it provides unique insights into the multifaceted nature of public sentiment and information dissemination during crises.

Social media platforms generate billions of dollars in revenue from U.S. youth: Findings from a simulated revenue model.

Social media platforms are suspected to derive hefty profits from youth users who may be vulnerable to negative mental health outcomes, including depression, anxiety, and eating disorders. Platforms, however, are not required to make these data publicly available, which may limit the abilities of researchers and policymakers to adequately investigate and regulate platform practices. This study aimed to estimate the number of U.S.-based child (0-12 years old) and adolescent (13-17 years old) users and the annual advertising revenue generated from youth across six major platforms. Data were drawn from public survey and market research sources conducted in 2021 and 2022. A simulation analysis was conducted to derive estimates of the number of users and the annual advertising revenue per age group and overall (ages 0-17 years) for 2022. The findings reveal that, across six major social media platforms, the 2022 annual advertising revenue from youth users ages 0-17 years is nearly $11 billion. Approximately 30-40% of the advertising revenue generated from three social media platforms is attributable to young people. Our findings highlight the need for greater transparency from social media platforms as well as regulation of potentially harmful advertising practices that may exploit vulnerable child and adolescent social media users.