L’expression de l’opinion

Users of forums, social networks and news portals now have the opportunity to publicly express their opinions on current political events, social issues, or their everyday lives. The analysis of opinion expression, which primarily represented a research topic in the field of language learning, has now become an important research challenge in the field of computational linguistics, which provides relevant solutions for various companies and organizations. The aim of this article is to analyse messages by which users of the social network Twitter reacted to an incident in which Emmanuel Macron was slapped in the face by a man as he went out to meet the public. We analysed the tweets that express agreement, disagreement and a neutral attitude towards the action. The analysis includes 80 tweets and refers to the textual, syntactic and lexical levels. The results show that tweets expressing disagreement have a typical declarative or exclamatory form, simple sentence structure and include explicit vocabulary expressing the author’s opinion (shameful, disrespectful). Tweets demonstrating agreement are more likely to have an exclamatory form, simple sentence structure and include an explicit term (well done, deserve a slap). Opinion-neutral tweets, on the other hand, are more likely to be formulated as declarative sentences with complex sentence structure and do not include an explicit term expressing the author’s opinion. The presented method is established on basic grammatical criteria (number of sentences, sentence structure, sentence form, keywords), which can also be applied to computational analysis of large collections of texts. In the future, the presented model could be applied to investigate various political, societal or healthcare challenges (elections, corruption or pandemic issues).

Cultural Influences on Opinion Expression in an Online and Offline Context

This study compares the expression of opinion in incongruent offline and online settings regarding the issue of gender desegregation in Kuwait’s public schools. Spiral of silence theory provides the theoretical foundation for examining the impact of certain cultural factors and religious influences on the expression of opinion, their relationship to the fundamental tenets of the theory, such as fear of social isolation, and Twitter use variables among respondents to a survey. The results to a questionnaire administered to 534 public and private university students indicate greater overall expression of opinion in the offline than online context. Offline and online, the nonconformist personality variable was a positive predictor of expression of opinion, and fear of social isolation was a negative predictor. The perceived position of Islam on the issue was a predictor of expression of opinion only in the offline context. Finally, daily average use of Twitter was an additional predictor of expression of opinion in the online environment.

Online Opinion Expression about Women Serving as Judges among University Students in Egypt and Kuwait: An Integrative Study of the Spiral of Silence and Uses and Gratifications

This study integrates the Spiral of Silence and Uses and Gratifications theories to examine the willingness of university students to express on Twitter their opinions about a controversial issue, women serving as judges in Kuwait and Egypt. The analysis of a survey of 640 respondents showed that they used Twitter for information seeking, opinion formation, opinion reinforcement, and social utility in discussions, and for its democratizing capability. Democratization was the only motive to predict the expression of opinion online. When the Kuwaiti and Egyptian samples were analyzed separately, the democratization motive predicted opinion expression for the Kuwaiti students but not for the Egyptian students. Interaction effects between motivations and size of the respondent's social network on Twitter were found to predict the online expression of opinion. For example, the variable assessing the size of a respondent's social network interacted with information seeking motivation and also with opinion reinforcement to predict opinion expression online.

Arabic Offensive and Hate Speech Detection Using a Cross-Corpora Multi-Task Learning Model

As social media platforms offer a medium for opinion expression, social phenomena such as hatred, offensive language, racism, and all forms of verbal violence have increased spectacularly. These behaviors do not affect specific countries, groups, or communities only, extending beyond these areas into people’s everyday lives. This study investigates offensive and hate speech on Arab social media to build an accurate offensive and hate speech detection system. More precisely, we develop a classification system for determining offensive and hate speech using a multi-task learning (MTL) model built on top of a pre-trained Arabic language model. We train the MTL model on the same task using cross-corpora representing a variation in the offensive and hate context to learn global and dataset-specific contextual representations. The developed MTL model showed a significant performance and outperformed existing models in the literature on three out of four datasets for Arabic offensive and hate speech detection tasks.

Online schadenfreude as an outcome of ideological polarization: a case in Hong Kong

PurposeThe impact of ideological polarization has been a serious concern, given its damages to society. In addition, Schadenfreude is increasingly common in the era of ideological polarization. Previous literature may have discussed the cause and outcomes of schadenfreude in general but not specifically related to ideological polarization. This study aims to serve to establish a more informed understanding of online schadenfreude as an outcome of ideological polarization and help society recover from the damages.Design/methodology/approachThe authors adopted a case research method with netnography for our study, given that the authors are exploring the phenomena of online schadenfreude, which involves multiple dimensions.FindingsThe authors identified a three-level model that illustrates how schadenfreude is driven as an outcome of ideological polarization, i.e. macro-environment, camp/partisan and target. These factors of different levels involve political viewpoint differences, perceived appearance, personal conduct, aggressive norms and polarized environment with a lack of conventional opinion expression channel. Moreover, attackers may demonstrate a belief in Karma, creativity and a sense of humor and may call for actions.Originality/valueWhile previous literature focused on the relationship between fake news, echo chambers and ideological polarization, this study is a relatively earlier one on studying schadenfreude as an outcome of ideological polarization, which would facilitate to formulate the solution to repair the damages created to ideological polarization. The authors also discussed the enablers as well as the self-reinforcing nature of ideological polarization, and provided some practical implications for politicians and government officials.