
Real-world events such as murders and political protests can trigger an increase in online hate speech directed against seemingly unrelated groups. The finding could help online moderators better predict when hateful content is most likely to be published, and what they should be looking out for, researchers say.