Anti-Semitism spikes on German social media after October attacks
The Surge of Anti-Semitism in German Social Media following the October 7 Attacks
Social and Information NetworksComputation and Language
Summary
After the Hamas attacks on Israel on October 7, 2023, anti-Semitic posts increased sharply on German social media sites Facebook and Telegram. The authors built a tool using large language models to detect 26 types of anti-Semitic content in over 125,000 posts from before and after the attacks. They found that Telegram had about ten times more anti-Semitic posts than Facebook, with each platform showing different common themes. The patterns of hate speech on the two platforms started to become more similar after the attack, with classic stereotypes rising on Facebook and Israel-related accusations growing on Telegram.
What this means in practice
- •For social media moderators: Improve content moderation systems on Facebook and Telegram with a model that detects detailed anti-Semitic categories using post text and user context to reduce false positives.
- •For law enforcement analysts: Use fine-grained detection of anti-Semitic language shifts after major events to better understand and respond to hate speech trends on different platforms.
Authors
Gregor Wiedemann, Daniel Wehrend
Abstract
We investigate the extent to which the Hamas attacks on Israel of October 7, 2023, have affected German social media debates about Judaism and Israel. For this, we develop an approach to detect 26 anti-Semitic categories in user postings via large language models (LLMs). The approach is applied to Facebook and Telegram posts (N=125,718) from three months before and after the event. Methodically, we test different open-weight models in two setups---with and without user information as additional context to the post text. The best setup achieves up to 83 % F1-score for binary anti-Semitism detection on our manually coded validation set. User context provides valuable information for most LLMs and drastically reduces false positives, for example, when (critically) reporting on anti-Semitic incidents. Concerning our topic, we find that anti-Semitism is surging significantly on both platforms, while being about ten times more prevalent on Telegram compared to Facebook. Facebook users express anti-Semitic views most likely in posts about an alleged genocide in Gaza carried out by the Israeli army, whereas classic anti-Semitic stereotypes related to power and conspiracy theories are dominant on Telegram. After the attack, the discourse patterns on both platforms show signs of convergence, as classic anti-Semitism increases on Facebook, whereas Israel-related categories surge on Telegram.