@inproceedings{WiegandRuppenhoferKleinbauer2019,
  author    = {Michael Wiegand and Josef Ruppenhofer and Thomas Kleinbauer},
  title     = {Detection of abusive language: the problem of biased datasets},
  series = {The 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Proceedings of the Conference Vol. 1. Minneapolis, Minnesota, June 2 - June 7, 2019},
  editor    = {Jill Burstein and Christy Doran and Thamar Solorio},
  publisher = {The Association for Computational Linguistics},
  address   = {Stroudsburg, PA, USA},
  isbn      = {978-1-950737-13-0},
  url       = {https://nbn-resolving.org/urn:nbn:de:bsz:mh39-90165},
  pages     = {602 -- 608},
  year      = {2019},
  abstract  = {We discuss the impact of data bias on abusive language detection. We show that classification scores on popular datasets reported in previous work are much lower under realistic settings in which this bias is reduced. Such biases are most notably observed on datasets that are created by focused sampling instead of random sampling. Datasets with a higher proportion of implicit abuse are more affected than datasets with a lower proportion.},
  language  = {en}
}