Search

Any form of communication that expresses hatred, prejudice, or hostility toward a particular individual or group of people based on attributes such as their race, religion, ethnicity, nationality, gender, sexual orientation, disability, or other protected characteristics is considered hate speech. Hate speech can be verbal, written, or symbolic. Hate speech can take many forms, and it often involves derogatory language, offensive stereotypes, or the incitement of violence or discrimination against the targeted individuals or groups. The content of hate speech is easy found in forum or discussion in social media include twitter. Twitter is a microblogging-based virtual entertainment where clients can peruse and compose text called tweets or tweets. This exploration executes order of disdain discourse in media Twitter utilizing IndoBERT. IndoBERT is the Indonesian form of BERT model utilizing over 220M words. It was a Convolutional Neural Network-based algorithm that had been modified. Th highlight extraction in Transformer isn't finished by convolution utilizing a part like CNN, however includes an encoder and decoder. The outcome demonstrates IndoBERT's excellent ability to categorize hate speech. [ABSTRACT FROM AUTHOR]

Social media platforms have been hailed as "politically disruptive communication technologies" (Hong & Nadler, 2012). Individuals express opinions and engage with politicians, the press, and each other on social media, sometimes using offensive language (Rossini et al., 2020). Content moderation has been adopted by many social media platforms to screen and evaluate offensive speech. In the present study we trained offensive speech classifiers to analyze offensive speech examples by integrating three archival datasets. We then used the trained classifier to examine a large body of comments about YouTube videos posted during the 2018 midterm election cycle. This provided information on the prevalence of various kinds of offensive comments and the pattern of content moderation used by YouTube. We also examined comment negativity using offensive speech lexicons. Our results showed systematic variance in the prevalence of offensive speech topics depending upon the political orientation of the content. Language use was significantly different between left and right‐leaning videos for comments related to sexism. [ABSTRACT FROM AUTHOR]

As algorithmic prediction systems have become widespread, fears that these systems may inadvertently discriminate against members of underrepresented populations have grown. With the goal of understanding fundamental principles that underpin the growing number of approaches to mitigating algorithmic discrimination, we investigate the role of information in fair prediction. A common strategy for decision-making uses a predictor to assign individuals a risk score; then, individuals are selected or rejected on the basis of this score. In this work, we study a formal framework for measuring the information content of predictors. Central to the framework is the notion of a refinement; intuitively, a refinement of a predictor z increases the overall informativeness of the predictions without losing the information already contained in z.We show that increasing information content through refinements improves the downstream selection rules across a wide range of fairness measures (e.g. true positive rates, false positive rates, selection rates). In turn, refinements provide a simple but effective tool for reducing disparity in treatment and impact without sacrificing the utility of the predictions. Our results suggest that in many applications, the perceived "cost of fairness" results from an information disparity across populations, and thus, may be avoided with improved information. [ABSTRACT FROM AUTHOR]

Through the sketch comedies "Ra Ra Show" and "Comedy Tonight" characterized by crossdressing male entertainers, drag debuted prominently into Singapore television during the 1990s. However, the former was taken off-air after complaints of "sexual innuendos" and non-standard colloquial English or Singlish. The latter, caricaturing the desexualized "wholesome" middle-age "auntie" continues to endear the public. Seemingly unavailable in both the institutional archives and online platforms in contrast to even the more controversial documentaries on LGBTQ topics, "Ra Ra Show" is no longer retrievable even as archives compared again to the relatively more easily available "Comedy Tonight." Comparing these case studies with the broader screening of gay-themed topics on Singapore's television, this paper draws out the trajectory of televisual queer texts from the stages of screenings to that of archiving. In turn it brings out the issues of accessibilities in both the viewing and subsequently ir/retrievability of television. [ABSTRACT FROM AUTHOR]

This article sets forth the problem of attitudes towards women in connection with their ethno-religious characteristics of the processing of their appearance (for example, hidjab (Islamic head-cover)) and sex and age of respondents. Research design. The object of the study were 794 people (350 male; 444 female), aged 18 to 40 years (mean age 27 years), residents of Rostov-on-Don, Russia. Methods. The study used the author's questionnaire, which consisted of stimulus material (2 sets of photos of 5 girls; in one case it was a portrait shot from the front with natural make-up, the second set - same girls, but in black hijab), respondents were asked to estimate the ratio of the girls in the photographs shown on a scale from friendly to hostile. Results. Based on the results we can say that there are hostile attitude towards the girl in hijab (based only on the design of the external appearance) and age and education level of respondence, but not in all measurements. Conclusions. So we can note that trigger mechanism of ethnolookism can be just a head cover, or other symbols of different cultures. Our findings on the line indicate that there is influence design appearance on the formation of relationships. It is concluded that the features of the assessment may depend on the age characteristics of the respondents and their level of education, but for the withdrawal of limits it is necessary to increase the sample by the age range. [ABSTRACT FROM AUTHOR]

Viewing Language as a marker of ethnicity, we use census data and a range of indices to examine language usage change in the United States over a 100 year period, 1910 to 2010. Our research is guided by three major questions. First, how much language diversity exists over the 100 year period? Second, how dominant is the English language from 1910 to 2010? And, third, what factors affect the odds of speaking a non-English language. In regard to the first question, we find that there is considerable language diversity over the 100 year period. However, only three languages, other than English, have survived in the top ten languages over the 100 year period: French, German, and Spanish. In regard to the second question, we find that English language dominance has declined over the 100 year period and that the biggest reason for this decline was the rise of the Spanish speaking population. In regard to the third question, we looked at it in two parts. In the first part, we examined location and find that increased residential segregation among foreign born enclaves has increased the relative numbers of non-English speakers in them over time. In the second part, we estimated a logistic regression model with foreign born status, foreign born enclaves and age as predictors for the years 1940 to 2010 and found that both foreign-born status and foreign-born enclaves greatly increased the odds of speaking a non-English language. While the impact of enclaves increased over time, as they simultaneously became more segregated. However, the most important determinant of speaking a non-English language was being foreign born. [ABSTRACT FROM AUTHOR]