German abusive language dataset with focus on COVID-19

The COVID-19 pandemic has had a significant impact on human lives globally. As a result, it is unsurprising that it has influenced hate speech and other sorts of abusive language on social media. Machine learning models have been designed to automatically detect such posts and messages, which necessitate a significant amount of labeled data. Despite the relevance of the COVID-19 topic in the field of abusive language detection, no annotated datasets with this focus are available. To solve these shortfalls, we target to create such a dataset. Our contributions are as follows: (1) a methodology for collecting abusive language data from Twitter with a substantial amount of abusive and hateful content, and (2) a German abusive language dataset with 4,960 annotated tweets centered on COVID-19. Both the methodology and the dataset are intended to aid researchers in improving abusive language detection.