User-generated Content In Social Media Research Articles

The Impact of Air Pollution Information on Individuals' Exercise Behavior: Empirical Study Using Wearable and Mobile Devices Data.

Physical exercise and exposure to air pollution have counteracting effects on individuals' health outcomes. Knowledge on individuals' real-time exercise behavior response to different pollution information sources remains inadequate. This study aims to examine the extent to which individuals avoid polluted air during exercise activities in response to different air pollution information sources. We used data on individuals' exercise behaviors captured by wearable and mobile devices in 83 Chinese cities over a 2-year time span. In our data set, 35.99% (5896/16,379) of individuals were female and 64% (10,483/16,379) were male, and their ages predominantly ranged from 18 to 50 years. We further augmented the exercise behavior data with air pollution information that included city-hourly level measures of the Air Quality Index and particulate matter 2.5 concentration (in µg/m3), and weather data that include city-hourly level measures of air temperature (ºC), dew point (ºC), wind speed (m/s), and wind direction (degrees). We used a linear panel fixed effect model to estimate individuals' exercise-aversion behaviors (ie, running exercise distance at individual-hour, city-hour, or city-day levels) and conducted robustness checks using the endogenous treatment effect model and regression discontinuity method. We examined if alternative air pollution information sources could moderate (ie, substitute or complement) the role of mainstream air pollution indicators. Our results show that individuals exhibit a reduction of running exercise behaviors by about 0.50 km (or 7.5%; P<.001) during instances of moderate to severe air pollution, and there is no evidence of reduced distances in instances of light air pollution. Furthermore, individuals' exercise-aversion behaviors in response to mainstream air pollution information are heightened by different alternative information sources, such as social connections and social media user-generated content about air pollution. Our results highlight the complementary role of different alternative information sources of air pollution in inducing individuals' aversion behaviors and the importance of using different information channels to increase public awareness beyond official air pollution alerts.

Assessing the EU Climate and Energy Policy Priorities for Transport and Mobility through the Analysis of User-Generated Social Media Content Based on Text-Mining Techniques

For over three decades, the European Union’s (EU) transport policy has aimed at fostering environmental sustainability and energy efficiency. Since 2015, European policymakers have focused more on three key sustainable development goals: decarbonizing the transport system, promoting low-emission mobility solutions, and transitioning to renewable and alternative fuels. To effectively communicate priorities and engage stakeholders, EU policymakers regularly use social media platforms like Twitter (now known as X). This active discourse involves policymakers, industrial stakeholders, the media, and the public, offering insights into the role of transport policy in addressing climate change and energy transition challenges. The current research endeavors to track and analyze the evolution of user-generated content related to climate change, energy transition, and smart mobility on Twitter from 2011 to 2021. This research uses text-mining and social network analysis techniques to quantitatively and qualitatively assess the dynamics of relevant EU policies and their effects. The study’s findings can be used to establish a robust monitoring and evaluation framework at the EU and national levels. This framework will assess the effectiveness of communicating strategic priorities for sustainable transport development. It also holds potential for application in other sectors, broadening its impact.

Performatively Archiving the Early Web: One Terabyte of Kilobyte Age

The abrupt closing of the web hosting service GeoCities – one of the most popular websites inhabited by users in the 90s – is a well-known example of the importance of web archiving for saving digital cultural heritage. GeoCities’ shutdown is a warning for today’s social media user-generated content that might suffer the same fate. Various web archiving organisations archived and nowadays present GeoCities pages. Olia Lialina and Dragan Espenschied engage with GeoCities’ legacy in their project One Terabyte of Kilobyte Age (2010–ongoing). This paper expands on the archival strategies of the artists and places their practice in the context of web archiving organisations – specifically Archive Team, the Internet Archive, and Restorativland – to understand how an artistic position may open up other ways of engaging with digital cultural heritage. This paper considers the archival practices of the organisations, artists, and users as a network of care, enabling different forms of remembrance. Whereas the archiving organisations preserve and present GeoCities statically, Lialina and Espenschied take a performative archival approach, in which they re-perform the dataset with old and new users. Through circulation and narration, One Terabyte of Kilobyte Age builds on the memories of the artists and users and aims to give the GeoCities heritage back to users. The project invites users to interpret and make meaning of GeoCities and embraces the fluidity of digital cultur e, whilst embodying the future archival insecurity of commercial platforms.

Detecting Mental Distresses Using Social Behavior Analysis in the Context of COVID-19: A Survey

Online social media provides a channel for monitoring people’s social behaviors from which to infer and detect their mental distresses. During the COVID-19 pandemic, online social networks were increasingly used to express opinions, views, and moods due to the restrictions on physical activities and in-person meetings, leading to a significant amount of diverse user-generated social media content. This offers a unique opportunity to examine how COVID-19 changed global behaviors regarding its ramifications on mental well-being. In this article, we surveyed the literature on social media analysis for the detection of mental distress, with a special emphasis on the studies published since the COVID-19 outbreak. We analyze relevant research and its characteristics and propose new approaches to organizing the large amount of studies arising from this emerging research area, thus drawing new views, insights, and knowledge for interested communities. Specifically, we first classify the studies in terms of feature extraction types, language usage patterns, aesthetic preferences, and online behaviors. We then explored various methods (including machine learning and deep learning techniques) for detecting mental health problems. Building upon the in-depth review, we present our findings and discuss future research directions and niche areas in detecting mental health problems using social media data. We also elaborate on the challenges of this fast-growing research area, such as technical issues in deploying such systems at scale as well as privacy and ethical concerns.

User-Generated Social Media Content in Consensual Conflict Resolution and Peacebuilding in Nigeria: A Case Study of Southern Kaduna, Kaduna State

User-Generated Social Media Content in Consensual Conflict Resolution and Peacebuilding in Nigeria: A Case Study of Southern Kaduna, Kaduna State

User-Generated Content in Social Media: A Twenty-Year Bibliometric Analysis in Hospitality

This article aims to present a bibliometric analysis regarding social media platforms and User-Generated Content (UGC) in hospitality. One hundred fifty-one peer-reviewed articles were analyzed using Webster’s and Watson’s (2002) methodology, a concept-driven methodology that helps analyze different concepts and contexts of a research field. Articles classified into five areas and a bibliometric analysis were presented to discuss the publication year, journals and publishers, authors, number of citations, research method implemented, social networking and users’ perceived value, user-generated content and travel planning, e-Word-of-Mouth (e-WOM) and brand image building, and hotel performance. The findings of this study showed that the number of studies in this field has increased over the last decade. However, exploration of the subject needs to be promoted (particularly experimental) because research in hospitality social media is still in the early phases on the grounds that publications concentrate on explicit subjects, regions, and sources of publication.

Improving Rumor Detection by Image Captioning and Multi-Cell Bi-RNN With Self-Attention in Social Networks

User-generated contents in social media are not verified before being posted. They could bring many problems if they were misused. Among various types of rumors, the authors focus on the type in which there's mismatch between images and their surrounding texts. They can be detected by multimodal feature fusion in RNNs with attention mechanism, but the relations between images and texts are not well-addressed. In this paper, the authors propose to improve rumor detection by image captioning and RNNs with self-attention. Firstly, they utilize the idea of image captioning to translate images into the corresponding text descriptions. Secondly, these caption words are represented by word embedding models and aggregated with surrounding texts using early fusion. Finally, multi-cell bi-directional RNNs with self-attention are used to learn important features to identify rumors. From the experimental results, the best F-measure of 0.882 can be obtained, which shows the potential of our proposed approach to rumor detection. Further investigation is needed for data in larger scale.

Social media user-generated content, online search traffic and offline car sales

PurposeThis study explores the direct relationship between social media user-generated content (UGC), online search traffic and offline light vehicle sales of different models.Design/methodology/approachThe long-run equilibrium relationship and short-run dynamic effects between the valence and volume of UGC, online search traffic and offline car sales are analyzed by applying the autoregressive distribution lag (ARDL) model.FindingsThe study found the following. (1) In the long-run relationship, the valence of online reviews on social media platforms is significantly negatively correlated with the sales of all models. However, in the short-run, the valence of online reviews has a significant positive correlation with all models in different lag periods. (2) The volume of online reviews is significantly positively correlated with the sales of all models in the long run. However, in the short run, the relationship between the volume of online reviews and the sales of lower-sales-volume cars is uncertain. There is a significant positive correlation between the volume of reviews and the sales of higher-sales-volume cars. (3) Online search traffic has a significantly negative correlation with the sales of all models in the long run. However, in the short run, there is no consistent conclusion on the relationship between online search traffic and car sales.Originality/valueThis study provides a reference for managers to use in their efforts to improve offline high-involvement product sales using online information.

Effects of social media brand-related content on fashion products buying behaviour – a moderated mediation model

PurposeThis study aims to examine in which circumstances consumer’s self-congruity moderates the indirect influence of consumer-based brand equity (mediating role) in the relationship between firm-created and user-generated social media content and intention to purchase fashion products.Design/methodology/approachIn this study, the authors carried out an online survey with social media users of fashion brands and collected data from 622 participants across two samples to investigate whether consumers’ perceptions of equity of fashion brands mediate the relationship between social media brand-related communication created by both firms and users and the intention to buy the fashion brands. The indirect relationship is further moderated by self-congruity.FindingsThe results indicate that brand equity mediates the relationship between social media communication and purchase intentions of fashion products, and self-congruity moderates the relationship between social media communication types and purchase intentions, such that higher/lower levels of self-congruity strengthen/weaken the impact of social media communication on purchase intentions.Originality/valueThis study contributes to the business and marketing literature by exploring how social media communication, branding and fashion align with the individual’s self-concept and buying behaviour.

Supply chain insights from social media users’ responses to panic buying during COVID-19: the herd mentality

PurposeThis study applied the herd mentality theory to explore local and global social media users’ responses to panic buying across the USA, UK and Australia during the COVID-19 crisis to understand the implications on operations and supply chains.Design/methodology/approachA total of 208,806 social media user-generated content (UGC) pieces were collected from Twitter in three countries – the USA, UK and Australia. The analysis of this big qualitative data was performed using machine learning–based software – Leximancer.FindingsPositive and negative sentiment towards panic buying during the COVID-19 crisis was observed in the UGC. No significant differences in social media UGC sentiment between the three countries were found; however, differences did exist in key themes. This suggests that the focus, not the sentiment, of consumers’ responses to panic buying differed across countries. Social media users follow their location-based and topic-consonant social “herd”, rather than the global “herd”.Research limitations/implicationsThis study was the first to show that social media users’ herd mentality differs in a crisis. The herd mentality of social networks is dependent on factors such as the geographic location of the social network (herd), which can differ from the global herd’s reaction, specifically in terms of topics evident in UGC.Practical implicationsOperations and supply chain managers need to include social media UGC analysis in their strategies in crisis management responses. The topics, not the sentiment, of consumers’ responses to panic buying require managerial actions.Originality/valueThis is the first study to show that herd mentality during a crisis, such as COVID-19, is not unidimensional and varies according to the location of the social media network with profound implications for operations and supply chain managers.

Psychiatry on Twitter: Content Analysis of the Use of Psychiatric Terms in French.

BackgroundWith the advent of digital technology and specifically user-generated contents in social media, new ways emerged for studying possible stigma of people in relation with mental health. Several pieces of work studied the discourse conveyed about psychiatric pathologies on Twitter considering mostly tweets in English and a limited number of psychiatric disorders terms. This paper proposes the first study to analyze the use of a wide range of psychiatric terms in tweets in French.ObjectiveOur aim is to study how generic, nosographic, and therapeutic psychiatric terms are used on Twitter in French. More specifically, our study has 3 complementary goals: (1) to analyze the types of psychiatric word use (medical, misuse, or irrelevant), (2) to analyze the polarity conveyed in the tweets that use these terms (positive, negative, or neural), and (3) to compare the frequency of these terms to those observed in related work (mainly in English).MethodsOur study was conducted on a corpus of tweets in French posted from January 1, 2016, to December 31, 2018, and collected using dedicated keywords. The corpus was manually annotated by clinical psychiatrists following a multilayer annotation scheme that includes the type of word use and the opinion orientation of the tweet. A qualitative analysis was performed to measure the reliability of the produced manual annotation, and then a quantitative analysis was performed considering mainly term frequency in each layer and exploring the interactions between them.ResultsOne of the first results is a resource as an annotated dataset. The initial dataset is composed of 22,579 tweets in French containing at least one of the selected psychiatric terms. From this set, experts in psychiatry randomly annotated 3040 tweets that corresponded to the resource resulting from our work. The second result is the analysis of the annotations showing that terms are misused in 45.33% (1378/3040) of the tweets and that their associated polarity is negative in 86.21% (1188/1378) of the cases. When considering the 3 types of term use, 52.14% (1585/3040) of the tweets are associated with a negative polarity. Misused terms related to psychotic disorders (721/1300, 55.46%) were more frequent to those related to depression (15/280, 5.4%).ConclusionsSome psychiatric terms are misused in the corpora we studied, which is consistent with the results reported in related work in other languages. Thanks to the great diversity of studied terms, this work highlighted a disparity in the representations and ways of using psychiatric terms. Moreover, our study is important to help psychiatrists to be aware of the term use in new communication media such as social networks that are widely used. This study has the huge advantage to be reproducible thanks to the framework and guidelines we produced so that the study could be renewed in order to analyze the evolution of term usage. While the newly build dataset is a valuable resource for other analytical studies, it could also serve to train machine learning algorithms to automatically identify stigma in social media.

Social Network Analysis for User Interaction Analysis on Social Media Regarding E-Commerce Business

E-commerce business requires promotion in introducing its products. One of the media that can be used is social media. There is a lot of information provided on social media, one of which is User Generated Content (UGC). UGC is a user's track record on social media that can be seen by other users. Social media analysis is needed to see the pattern of interaction between the company and its customers from UGC which is widely spread on social media. This can be used as an insight for companies in helping product marketing on social media. The method used in analyzing the interaction pattern of UGC in social media is Social Network Analysis (SNA). Social network modeling can help e-commerce businesses to understand the interaction patterns that occur on social media. The findings in this study show that the social network that is superior is the interaction social network regarding Lazada. Research also shows the key players for each e-commerce.

#PrayForDad: Learning the Semantics Behind Why Social Media Users Disclose Health Information

User-generated content in social media is increasingly acknowledged as a rich resource for research into health problems. One particular area of interest is in the semantics individuals’ evoke because they can influence when health-related information is disclosed. While there have been multiple investigations into why self-disclose occurs, much less is known about when individuals choose to disclose information about other people (e.g., a relative), which is a significant privacy concern. In this paper, we introduce a novel framework to investigate how semantics influence disclosure routines for 34 health issues. This framework begins with a supervised classification model to distinguish tweets that communicate personal health issues from confounding concepts (e.g., metaphorical statements that include a health-related keyword). Next, we annotate tweets for each health issue with linguistic and psychological categories (e.g. social processes, affective processes and personal concerns). Then, we apply a non-negative matrix factorization over a health issue-by-language category space. Finally, the factorized basis space is leveraged to group health issues into natural aggregations based around how they are discussed. We evaluate this framework with four months of tweets (over 200 million) and show that certain semantics correspond with whom a health mention pertains to. Our findings show that health issues related with family members, high medical cost and social support (e.g., Alzheimer's Disease, cancer, and Down syndrome) lead to tweets that are more likely to disclose another individual's health status, while tweets with more benign health issues (e.g., allergy, arthritis, and bronchitis) with biological processes (e.g., health and ingestion) and negative emotions are more likely to contain self-disclosures.

Understanding Cognitive Styles from User-Generated Social Media Content

The linguistic analyses on easily accessible, user-generated social media content offers great opportunities to identify individual characteristics such as their cognitive styles. In this paper, We explore the potential to use social media content to identify individuals' cognitive styles. We first employed crowdsourcing to collect Twitter users' cognitive styles using standard psychometric instruments. Then, we extracted the linguistic features of their social media postings. Leveraging these features, we build prediction models that provide estimates of cognitive styles through statistical regression and classification. We find that user generated content in social media provide useful information for characterizing people's cognitive styles. The models' performance indicates that the cognitive styles automatically inferred from social media are good proxies for the ground truth, and hence provides a promising and scalable way to automatically identify a large number of people's cognitive styles without reaching them individually.

Cultural Analytics of Large Datasets from Flickr

Deluge became a metaphor to describe the amount of information to which we are subjected, and very often we feel we are drowning while our access to information is rising. Devising mechanisms for exploring massive image sets according to perceptual attributes is still a challenge, even more when dealing with user-generated social media content. Such images tend to be heterogenous, and using metadata-only can be misleading. This paper describes a set of tools designed to analyze large sets of user-created art related images using image features describing color, texture, composition and orientation. The proposed pipeline permits to discriminate Flickr groups in terms of feature vectors and clustering parameters. The algorithms are general enough to be applied to other domains in which the main question is about the variability of the images.

Social media users’ crisis response: A lexical exploration of social media content in an international sport crisis

The topic of crisis communication and brand reputation management has attracted substantial academic attention, however minimal critique has been directed to the online response of social media users, through user-generated crisis-related communication. Drawing on crisis communication theory, we examine online sentiment of the social media public regarding the 2018 Cricket Australia ball tampering crisis over a 12-month period. It was found that sentiment towards the brand of Cricket Australia, shifted dramatically over this time. Social media users’ response progressed through four predominant stages: (1) framing and identity development; (2) sense making; (3) accountability; and, (4) vigilant rebuilding. These stages are depicted in the Social Media Users’ Crisis Response (SMUCR) framework presented in this paper. For brands involved in crisis situations, public response and social media user-generated content should be considered when formulating their communication and recovery strategies.

Is user-generated social media content useful for informing planning and management of emergency events? – An investigation of an active shooting event in a U.S. Airport

Terrorist attacks, and other short-scale emergencies such as active shooting events, have the potential to cause major loss of life they occur in passenger transportation terminals. Regulations often require airport authorities to be prepared for these events, but there is a lack of empirical data available to enhance response plans. Recently, there has been significant interest on how user-generated content available on social media platforms could aid with management and planning in response to emergencies. However, little of this has focused on active shooting events in particular. This paper uses a case study approach to analyse the content of social media posts to evaluate their potential to aid with situational awareness during the response phase of short-scale emergency events. Alongside a literature review conducted into this topic, this paper details the results of the analysis of 131,600 Twitter posts related to the case study active shooting event – which occurred in 2017 at Fort-Lauderdale Hollywood airport in Florida. Qualitative content analysis, and two machine learning algorithms – a Support Vector Machine, and a Latent Dirichlet Allocation model were used to categorise the tweets produced in relation to this event. Results indicate that the majority of tweets produced either simply stated that the event had occurred, or were emotional in nature, expressing worry or prayers for victims. Very few posts detailing behavioural responses of individuals involved in the event were identified. Thus, it is suggested that social media data would not be of particular use to improving situational awareness during response to short-scale emergency events in transportation terminals. Instead, social media may be of more use for information dissemination or conversations and coordination purposes during the preparation, recovery or mitigation phases of an emergency event.

How to address data privacy concerns when using social media data in conservation science.

Social media data are being increasingly used in conservation science to study human-nature interactions. User-generated content, such as images, video, text, and audio, and the associated metadata can be used to assess such interactions. A number of social media platforms provide free access to user-generated social media content. However, similar to any research involving people, scientific investigations based on social media data require compliance with highest standards of data privacy and data protection, even when data are publicly available. Should social media data be misused, the risks to individual users' privacy and well-being can be substantial. We investigated the legal basis for using social media data while ensuring data subjects' rights through a case study based on the European Union's General Data Protection Regulation. The risks associated with using social media data in research include accidental and purposeful misidentification that has the potential to cause psychological or physical harm to an identified person. To collect, store, protect, share, and manage social media data in a way that prevents potential risks to users involved, one should minimize data, anonymize data, and follow strict data management procedure. Risk-based approaches, such as a data privacy impact assessment, can be used to identify and minimize privacy risks to social media users, to demonstrate accountability and to comply with data protection legislation. We recommend that conservation scientists carefully consider our recommendations in devising their research objectives so as to facilitate responsible use of social media data in conservation science research, for example, in conservation culturomics and investigations of illegal wildlife trade online.

Depression and anorexia detection in social media as a one-class classification problem

Taking advantage of the increasing amount of user-generated content in social media, some computational methods have already been proposed for detecting people suffering from depression and anorexia. Such complex tasks have been tackled as a binary classification problem using, in most cases, automatically generated training data. Despite its promising results, this approach has some important drawbacks, namely: it suffers from a severely skewed class distribution, the negative class is very diverse since it attempts to model all kinds of healthy users, and, above all, there is not a complete certainty about annotations, especially for the negative cases (i.e., healthy users). Motivated by these issues, in this paper, we propose to face the detection of these disorders following a one-class classification (OCC) approach. Particularly, we introduce two new instance-based OCC methods especially suited to manage the high diversity of content from social media documents. Taking up ideas from the gravitational attraction force, these methods evaluate the relation of documents by their strengths, considering their distances as well as their masses (relevance) with respect to the target task. Experiments were conducted on depression and anorexia benchmark datasets. The obtained results are encouraging; the overall performance was better than the results from other standard OCC methods, and competitive with regard to state-of-the-art results from binary classification approaches.

Unveiling the Hidden Truth of Drug Addiction: A Social Media Approach Using Similarity Network-Based Deep Learning

ABSTRACT Opioid use disorder (OUD) is an epidemic that costs the U.S. healthcare systems $504 billion annually and poses grave mortality risks. Existing studies investigated OUD treatment barriers via surveys as a means to mitigate this opioid crisis. However, the response rate of these surveys is low due to social stigma around opioids. We explore user-generated content in social media as a new data source to study OUD. We design a novel IT system, SImilarity Network-based DEep Learning (SINDEL), to discover OUD treatment barriers from patient narratives and address the challenge of morphs. SINDEL significantly outperforms state-of-the-art NLP models, reaching an F1 score of 76.79 percent. Thirteen types of treatment barriers were identified and verified by domain experts. This work contributes to information systems with a novel deep-learning-based approach for text analytics and generalized design principles for social media analytics methods. We also unveil the hurdles patients endure during the opioid epidemic.