Search results

The purpose of this paper is to design organization message content strategies and analyse their information diffusion on the microblogging website, Twitter.

Using data from 29 brands and 9392 tweets, message strategies on twitter are classified into four strategies. Using content analysis all the tweets are classified into informational strategy, transformational strategy, interactional strategy and promotional strategy. Additionally, the information diffusion for the developed message strategies was explored. Furthermore, message content features such as text readability features, language features, Twitter-specific features, vividness features on information diffusion are analysed across message strategies. Additionally, the interaction between message strategies and message features was carried out.

Finding reveals that informational strategies were the dominant message strategy on Twitter. The influence of text readability features language features, Twitter-specific features, vividness features that influenced information diffusion varied across four message strategies.

This study offers a completely novel way for effectively analysing information diffusion for branded tweets on Twitter and can show a path to both researchers and practitioners for the development of successful social media marketing strategies.

Getting high-quality data by removing the noisy data from the user-generated content (UGC) is the first step toward data mining and effective decision-making based on ubiquitous and unstructured social media data. This paper aims to design a framework for revoking noisy data from UGC.

In this paper, the authors consider a classification-based framework to remove the noise from the unstructured UGC in social media community. They treat the noise as the concerned topic non-relevant messages and apply a text classification-based approach to remove the noise. They introduce a domain lexicon to help identify the concerned topic from noise and compare the performance of several classification algorithms combined with different feature selection methods.

Experimental results based on a Chinese stock forum show that 84.9 per cent of all the noise data from the UGC could be removed with little valuable information loss. The support vector machines classifier combined with information gain feature extraction model is the best choice for this system. With longer messages getting better classification performance, it has been found that the length of messages affects the system performance.

The proposed method could be used for preprocessing in text mining and new knowledge discovery from the big data.

Distinguishing phishing emails from legitimate emails continues to be a difficult task for most individuals. This study aims to investigate the psycholinguistic factors associated with deception in phishing email text and their effect on end-user ability to discriminate phishing emails from legitimate emails.

Email messages and end-user decisions collected from a laboratory phishing study were validated and analyzed using natural language processing methods (Linguistic Inquiry Word Count) and penalized regression models (LASSO and Elastic Net) to determine the linguistic dimensions that attackers may use in phishing emails to deceive end-users and measure the impact of such choices on end-user susceptibility to phishing.

We found that most participants, who played the role of a phisher in the study, chose to deceive their end-user targets by pretending to be a familiar individual and presenting time pressure or deadlines. Results show that use of words conveying certainty (e.g. always, never) and work-related features in the phishing messages predicted higher end-user vulnerability. On the contrary, use of words that convey achievement (e.g. earn, win) or reward (cash, money) in the phishing messages predicted lower end-user vulnerability because such features are usually observed in scam-like messages.

Insights from this research show that analyzing emails for psycholinguistic features associated with computer-mediated deception could be used to fine-tune and improve spam and phishing detection technologies. This research also informs the kinds of phishing attacks that must be prioritized in antiphishing training programs.

Applying natural language processing and statistical modeling methods to analyze results from a laboratory phishing experiment to understand deception from both attacker and end-user is novel. Furthermore, results from this work advance our understanding of the linguistic factors associated with deception in phishing email text and its impact on end-user susceptibility.

The authors examine the contemporaneous and causal association between tweet features (bullishness, message volume and investor agreement) and market features (stock returns, trading volume and volatility) using 140 South African companies and a dataset of firm-level Twitter messages extracted from Bloomberg for the period 1 January 2015 to 31 March 2020.

Panel regressions with ticker fixed-effects are used to examine the contemporaneous link between tweet features and market features. To examine the link between the magnitude of tweet features and stock market features, the study uses quantile regression.

No monotonic relationship is found between the magnitude of tweet features and the magnitude of market features. The authors find no evidence that past values of tweet features can predict forthcoming stock returns using daily data while weekly and monthly data shows that past values of tweet features contain useful information that can predict the future values of stock returns.

The study is among the earlier to examine the association between textual sentiment from social media and market features in a South African context. The exploration of the relationship across the distribution of the stock market features gives new insights away from the traditional approaches which investigate the relationship at the mean.

In the field of public relations and communication management, message evaluation has been one of the starting points for evaluation and measurement research at least since the 1970s. Reliable and valid message evaluation has a central role in message effects research and campaign design in other disciplines as well as communication science. The purpose of this paper is to offer a message testing protocol to efficiently acquire valid and reliable message evaluation data.

A message testing protocol is described in terms of how to conceptualize and evaluate the content and format of messages, in terms of procedures for acquiring and testing messages and in terms of using efficient, reliable and valid measures of perceived message effectiveness (PME) and perceived argument strength (PAS). The evidence supporting the reliability and validity of PME and PAS measures is reviewed.

The message testing protocol developed and reported is an efficient, reliable and valid approach for testing large numbers of messages.

Researchers’ ability to select candidate messages for subsequent deeper testing, for various types of communication campaigns, and for research in theory testing contexts is facilitated. Avoiding the limitations of using a single instance of a message to represent a category (also known as the case-category confound) is reduced.

Communication campaign designers are armed with tools to assess messages and campaign concepts quickly and efficiently, reducing pre-testing time and resources while identifying “best-in-show” examples and prototypes.

Message structures are conceptualized in terms of content and format features using theoretically driven constructs. Measures of PAS and PME are reviewed for their reliability, construct and predictive validity, finding that the measures are acceptable surrogates for actual effectiveness for a wide variety of messages and applications. Coupled with procedures that reduce confounding by randomly nesting messages within respondents and respondents to messages, the measures used and protocol deployed offer an efficient and utilitarian approach to message testing and modeling.

This study investigates how media brand knowledge, defined as a structural feature of the message, influences emotional and attentional responses to, and memory of, news messages.

Self-reports, facial electromyography (EMG) and electroencephalography were used as indices of emotional valence, arousal and attention in response to 42 news messages, which varied along the valence and involvement dimensions and were framed with different media brands varying along the familiarity and credibility dimensions.

Compared to the no-brand condition, news framed with brands elicited more attention. The memory tests indicated that strong media brands override the effect of involvement in information encoding, whereas details of news presented with Facebook were not well encoded. However, the headlines of news framed with Facebook were well retrieved. In addition, negative and high-involvement news elicited higher arousal ratings and corrugator EMG activity. News framed with familiar and high-credibility brands elicited higher arousal ratings.

Relevant for both brand managers and audiences, the findings show that building credibility and familiarity both work as brand attributes to differentiate media brands and influence information processing.

The results highlight the importance of media brands in news reading: as a structural feature, the brand is used as a proxy to process the message content. The study contributes by investigating how the type of source influences the reception and encoding of the mediated information; by investigating the emotional effects of brands; and by confirming previous findings in media psychology literature.

Twitter is a popular microblogging service which has proven, in recent years, its potential for propagating news and information about developing events. The purpose of this paper is to focus on the analysis of information credibility on Twitter. The purpose of our research is to establish if an automatic discovery process of relevant and credible news events can be achieved.

The paper follows a supervised learning approach for the task of automatic classification of credible news events. A first classifier decides if an information cascade corresponds to a newsworthy event. Then a second classifier decides if this cascade can be considered credible or not. The paper undertakes this effort training over a significant amount of labeled data, obtained using crowdsourcing tools. The paper validates these classifiers under two settings: the first, a sample of automatically detected Twitter “trends” in English, and second, the paper tests how well this model transfers to Twitter topics in Spanish, automatically detected during a natural disaster.

There are measurable differences in the way microblog messages propagate. The paper shows that these differences are related to the newsworthiness and credibility of the information conveyed, and describes features that are effective for classifying information automatically as credible or not credible.

The paper first tests the approach under normal conditions, and then the paper extends the findings to a disaster management situation, where many news and rumors arise. Additionally, by analyzing the transfer of our classifiers across languages, the paper is able to look more deeply into which topic-features are more relevant for credibility assessment. To the best of our knowledge, this is the first paper that studies the power of prediction of social media for information credibility, considering model transfer into time-sensitive and language-sensitive contexts.

The purpose of this paper is to report findings from an interpretive case study on user participation in the open source software (OSS) development context.

Through an empirical, interpretive case study and a literature review utilising the metaphor of text in the analysis, this paper provides a refined conceptualisation of user participation in OSS development.

The paper reveals that different kinds of meanings have been attached to users and to their participation. User participation is both direct and indirect in the OSS development context. Some user groups actively take part in OSS development, while others are merely represented in it. Different kinds of intermediaries “representing the users” are identified.

The research is based on one case study on a small but active OSS project with an interest in users. Other kinds of OSS projects should be analysed. The analysis was focused on a discussion forum, but users can take part in OSS development by other means as well. Paths for future work should include the gathering of more varied empirical data.

The findings indicate that users can provide feedback to the development through discussion forums in the distributed environment, but there is a need to support the users in doing so and the developers in analysing the data. The importance of different kinds of intermediaries “representing the users” is highlighted.

The paper provides thorough empirical insights and a refined conceptualisation of user participation addressing the currently weakly empirically explored OSS development context.

To better understand executive communication on social media, the purpose of this paper is to examine the pattern of messages posted by chief executive officers (CEOs) on Twitter and their retweetability (rate of reposting by other users).

The study data comprises 1,068 original tweets randomly selected from all Fortune 1000 CEOs’ tweets in 2014. The impact of the contextual factors (industry background, activeness, and Twitter age) and content factors (content types, supplementary information, and linguistic features) on retweetability was examined through regression analyses.

CEOs tweet to share information and insights, to promote their companies or products, to update work or life status, and to interact with the public. Original insights, promotional messages, and seasonal greetings were most likely to be retweeted. CEOs’ backgrounds, usage of hashtags, and certainty of language were also positively associated with retweetability.

CEOs may enhance their online social influence through demonstrating leadership by sharing insights about their organization or industry and posting topical messages (e.g. season’s greetings). Furthermore, CEOs could use hashtags strategically to initiate or participate in discussions and promote their personal visibility.

This study is one of the first to evaluate how leaders of the largest companies in the USA communicate on Twitter. It contributes to a theoretical understanding of the factors underlying online influence – the influence of the status of the online communicator vs the message content on information forwarding.

This paper aims to examine the effect of cross‐channel integration of an advertiser's television spot that invited viewers to play an online game and web site that featured the game on consumers' perceived media engagement and brand attitudes. An important factor, personal involvement is also to be examined.

In total, 185 college students participate in an experimental study on participation and involvement. Participants fill out surveys and personal involvement is determined using various advertisements.

The results reveal that interaction effects are evident between cross‐channel integration of advertising and personal involvement on media engagement and brand attitudes.

This study is limited, investigating only one brand, one product category, and two media that can feature advertising messages regarding cross‐channel integration. Future studies should investigate the cross‐channel integration effect on different segments of target consumers.

Advertisers should consider enhancing their media strategies. In particular, advertisers should manage a media plan that includes specific sets of contacts to enhance the effects of cross‐channel integration of advertising.

The paper helps substantiate the strategic value of cross‐channel integration of advertising on branding by assessing the value of cross‐channel integration of advertising.