Search results

In the age of social media, when publishers are vying for consumer attention, click-baits have become very common. Not only viral websites but also mainstream publishers, such as news channels, use click-baits for generating traffic. Therefore, click-bait detection and prediction of click-bait virality have become important challenges for social media platforms to keep the platform click-bait free and give a better user experience. The purpose of this study is to try exploring how the contents of the social media posts and the article can be used to explain and predict social media posts and the virality of a click-bait.

This study has used 17,745 tweets from Twitter with 4,370 click-baits from top 27 publishers and applied econometric along with machine learning methods to explain and predict click-baitiness and click-bait virality.

This study finds that language formality, readability, sentiment scores and proper noun usage of social media posts and various parts of the target article plays differential and important roles in click-baitiness and click-bait virality.

The paper contributes toward the literature of dark behavior in social media at large and click-bait prediction and explanation in particular. It focuses on the differential roles of the social media post, the article shared and the source in explaining click-baitiness and click-bait virality via psycho-linguistic framework. The paper also provides explanability to the econometric and machine learning predictive models, thus performing methodological contribution too.

The paper helps social media managers create a mechanism to detect click-baits and also predict which ones of them can become viral so that corrective measures can be taken.

To the best of the authors’ knowledge, this is one of the first papers which focus on both explaining and predicting click-baitiness and click-bait virality.

The purpose of this paper is to conceptualize the use of Twitter hashtag as a strategy to enhance the visibility and symbolic power of social movement-related information. It examined how characteristics of hashtag drove information virality during a networked social movement.

Twitter data from two days during the Occupy Wall Street Movement in 2011 were collected. With network analysis, the authors identified popular hashtag types and examined hashtag co-occurrence patterns during the two contrasting movement days. It also provides a comparative analysis of how major types of viral hashtag may play different roles depending on different movement cycles.

The authors found that the role of hashtag influencing information virality may vary based on the context of the tweets. For example, movement participants applied more strategic hashtag combinations during the unexpected event day to reach different social circles. Consistent patterns were identified in mobilizing influential actors such as public figures. Different use patterns of media outlet hashtag were found across the two days.

Implications on how hashtag type and event dynamics may shape hashtag co-occurrence patterns were discussed.

This paper studies, based on the theory of service-dominant logic, the effect of value co-creation practices (linking and materializing) on engagement dimensions (popularity, commitment and virality). The main objective is to analyze the influence of value co-creation practices on engagement at international trade shows organizer association on Twitter.

This paper studies the usage of Twitter by the Specialty Food Association, which organizes one of the top five foods and beverage international trade show in the United States. To achieve the research objective, the authors have analyzed 1,608 posts on Twitter from the Twitter account @Specialty_Food. A content analysis was performed using Krippendorff's (2004) recommendations, and the data were analyzed using regression analysis with optimal scaling and Kruskal–Wallis Test.

According to the results, some materializing practices influence popularity, commitment, virality and global engagement on Twitter. While the usage of some linking practices influences respectively commitment and popularity.

These results provide valuable information for business-to-business (B2B) contexts and answer a research gap reported in previous literature, which affirms that more research is needed about the relationship between service systems and engagement. From a general view, to generate more engagement on social media in B2B contexts, it is recommended to prioritize posts that incorporate live and online events based on collaborative and dynamic human interactions, following by business ideas and business cases.

The purpose of this paper is to seek reasons for some videos going viral over YouTube (a type of social media platform).

Using YouTube APIs (Application Programming Interface) and Webometrics analyst tool, the authors collected data on about 100 all-time-most-viewed YouTube videos and information about the users associated with the videos. The authors constructed and tested an empirical model to understand the relationship among users’ social and non-social capital (e.g. User Age, Gender, View Count, Subscriber, Join Date, Total Videos Posted), video characteristics (Post Date, Duration, and Video Category), external network capital (in-links and hit counts), and Virality (Likes, Dislikes, Favorite Count, View Count, and Comment Count). Partial least square and Webometric analysis was used to explore the association among the constructs.

Among other findings, the results showed that popularity of the videos was not only the function of YouTube system per se, but that network dynamics (e.g. in-links and hits counts) and offline social capital (e.g. fan base and fame) play crucial roles in the viral phenomenon, particularly view count.

The authors for the first time constructed and tested an empirical model to find out the determinants of viral phenomenon over YouTube.

Social media platforms play a key role in information propagation and there is a need to study the same. This study aims to explore the impact of the number of close communities (represented by cliques), the size of these close communities and its impact on information virality.

This study identified 6,786 users from over 11 million tweets for analysis using sentiment mining and network science methods. Inferential analysis has also been established by introducing multiple regression analysis and path analysis.

Sentiments of content did not have a significant impact on the information virality. However, there exists a stagewise development relationship between communities of close friends, user reputation and information propagation through virality.

This paper contributes to the theory by introducing a stagewise progression model for influencers to manage and develop their social networks.

There is a gap in the existing literature on the role of the number and size of cliques on information propagation and virality. This study attempts to address this gap.

This paper aims to contribute to the emerging body of research on firestorms, specifically on the inflammatory user-generated content (UGC) created in response to brand transgressions. By analyzing and segmenting UGC created and shared in the wake of three different events, the authors identify which type of inflammatory message is most likely to be widely shared; thus, contributing to a possible online firestorm.

Tweets were collected involving brand transgressions in the retail, fast food and technology space from varying timeframe and diverse media coverage. Then, the tweets were coded for message intention and analyzed with linguistics software to determine the message characteristics and framing. A two-step cluster analysis identified three types of UGC.

The authors found that message dimensions and the framing of tweets in the context of brand transgressions differed in characteristics, sentiment, call to action and the extent to which the messages were shared. The findings contradict traditional negative word-of-mouth studies involving idiosyncratic service and product failure. During online brand firestorms, rational activism messages with a call to action, generated in response to a firm’s transgression or “sparks,” have a higher likelihood of being shared (virality).

This research provides novel insights into UGC created after brand transgressions. Different types of messages created after these events vary in the extent that they “fan the flames” of the transgression. A message typology and flowchart are provided to assist managers in identifying and responding to three message types: ash, sparks and embers.

Social media have increasingly gained credibility as information sources in emergencies. Retweeting or resharing nature has made Twitter a popular medium of information dissemination. The purpose of this article is to enhance our understanding of both linguistic style and content properties (i.e. both affective and informational contents) that drives resharing behavior or virality of disaster messages on Twitter. We investigate this issue in the context of natural disaster crisis.

In this study, the authors develop, drawing upon language expectancy and uncertainty reduction theories as an enabling framework, hypotheses about how the language (i.e. style and content) influence resharing behavior. They employ a natural language processing of disaster tweets to examine how the language – linguistic style (concrete and interactive language) and linguistic content (information- and affect-focused language) – affects resharing behavior on Twitter during natural disasters. To examine the effects of both linguistic style and content factors on virality, a series of negative binomial regressions were conducted, particularly owing to the highly skewed count data.

Our analysis of tweets from the 2013 Colorado floods shows that resharing disasters tweets increases with the use of concrete language style during acute emergencies. Interactive language is also positively associated with retweet frequency. In addition, neither positive nor negative emotional tweets drive down resharing during acute crises, while information-focused language content has a significantly positive effect on virality.

Agencies for public safety and disaster management or volunteer organizations involved in disseminating crisis and risk information to the public may leverage the impacts of the linguistic style and language content through the lens of our research model. The findings encourage practitioners to focus on the role of linguistic style cues during acute disasters. Specifically, from the uncertainty reduction perspective, using concrete language in the disaster tweets is the expected norm, leading to a higher likelihood of virality. Also, interactively frame disaster tweets are more likely to be diffused to a larger number of people on Twitter.

The language that people use offer important psychological cue to their intentions and motivations. However, the role of language on Twitter has largely been ignored in this crisis communication and few prior studies have examined the relationship between language and virality during acute emergencies. This article explains the complex and multifaceted nature of information resharing behavior using a multi-theoretical approach – including uncertainty reduction and language expectancy theory – to understand effects of language style and content cues on resharing behavior in the context of natural crisis events.

The purpose of this paper is to examine the factors that significantly affect the prediction of fake news from the virality theory perspective. The paper looks at a mix of emotion-driven content, sentimental resonance, topic modeling and linguistic features of news articles to predict the probability of fake news.

A data set of over 12,000 articles was chosen to develop a model for fake news detection. Machine learning algorithms and natural language processing techniques were used to handle big data with efficiency. Lexicon-based emotion analysis provided eight kinds of emotions used in the article text. The cluster of topics was extracted using topic modeling (five topics), while sentiment analysis provided the resonance between the title and the text. Linguistic features were added to the coding outcomes to develop a logistic regression predictive model for testing the significant variables. Other machine learning algorithms were also executed and compared.

The results revealed that positive emotions in a text lower the probability of news being fake. It was also found that sensational content like illegal activities and crime-related content were associated with fake news. The news title and the text exhibiting similar sentiments were found to be having lower chances of being fake. News titles with more words and content with fewer words were found to impact fake news detection significantly.

Several systems and social media platforms today are trying to implement fake news detection methods to filter the content. This research provides exciting parameters from a viral theory perspective that could help develop automated fake news detectors.

While several studies have explored fake news detection, this study uses a new perspective on viral theory. It also introduces new parameters like sentimental resonance that could help predict fake news. This study deals with an extensive data set and uses advanced natural language processing to automate the coding techniques in developing the prediction model.

YouTube allows its users to upload and view videos on its platform. YouTube provides notification to the subscribers whenever a channel uploads a new video thereby making the channel subscribers the potential viewers of the video. And thus, they are the first to come to know about any new offering. But later on, the view count also increases due to virality, that is, mass sharing of the content by the users on different social media platforms similar to word-of-mouth in the field of marketing. Therefore, the purpose of this paper is to examine different diffusion patterns as they can help to inflate traffic and generate revenue.

YouTube's view count grows majorly through virality. The pattern of view count growth has generally been considered unimodal in most of the available research in the field of YouTube. In the present work, the growth process due to views through the subscribers and views due to word-of-mouth (virality) is presented. Considering that the impact of virality in view count growth comes later in the video life cycle; the viewing patterns of both the segments have been mathematically modeled; independently.

Different models have been proposed to capture the view count growth pattern and how the impact of virality changes the view count growth curve and thereby results in a multimodal curve structure. The proposed models have been verified on various view count data sets of YouTube videos using SPSS (Statistical Package for the Social Sciences), and their ranks have been determined using a weighted criteria–based approach. The results obtained clearly depict the presence of many modes in the life cycle of view counts.

Till now, the literature is evident of the video life cycle following a bell shape curve. This study claims that the initial thrust is by subscribers and then the contribution in the view count by people watching via word-of-mouth comes into picture and brings in another hump in the growth curve.

The purpose of this study/paper is to propose a mathematical model that is able to predict the future popularity based on the view count of a particular YouTube video. Since the emergence of video-sharing sites from early 2005, YouTube has been pioneering in its performance and holds the largest share of internet traffic. YouTube plays a significant role in popularizing information on social network. For all social media sites, viewership is an important and vital component to measure diffusion on a video-sharing site, which is defined in terms of the number of view counts. In the era of social media marketing, companies demand an efficient system that can predict the popularity of video in advance. Diffusion prediction of video can help marketing firms and brand companies to inflate traffic and help the firms in generating revenue.

In the present work, viewership is studied as an important diffusion-affecting parameter pertaining to YouTube videos. Primarily, a mathematical diffusion model is proposed to predict YouTube video diffusion based on the varying situations of viewership. The proposal segregates the total number of viewers into two classes – neoterics viewers, i.e. viewers those viewing a video on a direct basis, and followers, i.e. viewers those watching under the influence.

The approach is supplemented with numerical illustration done on the real YouTube data set. Results prove that the proposed approach contributes significantly to predict viewership of video. The proposed model brings predicted viewership and its classification highly close to the true value.

Thereby, a behavioral rationale for the modeling and quantification is offered in terms of the two varied and yet connected classes of viewers – “neoterics” and “followers.”