Online Library

The research series Hate Speech and Radicalisation on the Internet provides interdisciplinary insights into the current developments of extremist activities on the internet. With the aid of expert contributions from all over Germany, the psychological, political, anthropological and technological aspects of online hate speech and radicalisation will be considered and recommendations will be made for political leaders, social media platforms as well as NGOs and activists.

The increasing presence of hate speech on social media has drawn significant investment from governments, companies, and empirical research. Existing methods typically use a supervised text classification approach that depends on carefully engineered features. However, it is unclear if these features contribute equally to the performance of such methods. We conduct a feature selection analysis in such a task using Twitter as a case study, and show findings that challenge conventional perception of the importance of manual feature engineering: automatic feature selection can drastically reduce the carefully engineered features by over 90% and selects predominantly generic features often used by many other language related tasks; nevertheless, the resulting models perform better using automatically selected features than carefully crafted task-specific features.

We use the construct of the “other” to explore how hate operates rhetorically within the virtual conclave of Stormfront, credited as the first hate Web site. Through the Internet, white supremacists create a rhetorical vision that resonates with those who feel marginalized by contemporary political, social, and economic forces. However, as compared to previous studies of on-line white supremacist rhetoric, we show that Stormfront discourse appears less virulent and more palatable to the naive reader. We suggest that Stormfront provides a “cyber transition” between traditional hate speech and “reasonable racism,” a tempered discourse that emphasizes pseudo- rational discussions of race, and subsequently may cast a wider net in attracting audiences.

Social media has become an essential part of the daily routines of children and adolescents. Moreover, enormous efforts have been made to ensure the psychological and emotional well-being of young users as well as their safety when interacting with various social media platforms. In this paper, we investigate the exposure of those users to inappropriate comments posted on YouTube videos targeting this demographic. We collected a large-scale dataset of approximately four million records, and studied the presence of five age-inappropriate categories and the amount of exposure to each category. Using natural language processing and machine learning techniques, we constructed ensemble classifiers that achieved high accuracy in detecting inappropriate comments. Our results show a large percentage of worrisome comments with inappropriate content: we found 11% of the comments on children’s videos to be toxic, highlighting the importance of monitoring comments, particularly on children platforms.

This report aims to provide an in-depth analysis on the state of extremism and hate on TikTok. It is the culmination of three months of research on a sample of 1,030 videos, equivalent to just over eight hours of content, posted on the social media platform. These videos were used to promote hatred, as well as glorify extremism and terrorism.

ISD set out to examine the state of hate and extremism on TikTok in two ways. The first objective involved analysing how individuals or groups promote hateful ideologies and target people on the platform based on numerous protected attributes such as ethnicity, religion, gender or others. Second, using the same framework, ISD investigated how features on TikTok like profiles, hashtags, share functions, video effects and music are used to spread hate.

This report seeks to start a conversation around how platforms like TikTok can improve their own practices to protect users from harm. Additionally, it underscores the clear need for independent oversight of such platforms, which currently leave users and the wider public open to significant risks to their health, security and rights.

— The reporting, recording and incidence of online hate speech has increased over the past two years.
— While the number of people personally targeted remains relatively low, large numbers of people are being exposed to online hate speech, potentially causing decreased life satisfaction. In particular, an increasingly large number of UK children (aged 12-15) report that they are exposed to hateful content online.
— Online hate speech tends to spike for 24-48 hours after key national or international events such as a terror attack, and then rapidly fall, although the baseline of online hate can remain elevated for several months. Where it reaches a certain level, online hate speech can translate into offline hate crime on the streets.
— Hate crime, including hate speech, is both hard to define and hard to prosecute. A patchwork of hate crime laws has developed over the last two decades, but there is concern the laws are not as effective as they could be, and may need to be streamlined and/or extended – for example to cover gender and age-related hate crime. The Law Commission is currently reviewing hate crime legislation, and has separately completed a preliminary review of the criminal law in relation to offensive
and abusive online communications, concluding there was “considerable scope for reform”.
— According to a recent survey by Demos, the public appreciates the difficult trade-off between tackling hate crime and protecting freedom of speech, with 32% in favour of a safety first approach, 23% in favour of protecting civil liberties, and 42% not favouring either option.