DALC: the Dutch Abusive Language Corpus

Tommaso Caselli, Arjan Schelhaas, Marieke Weultjes, Folkert Leistra, Hylke van der Veen, Gerben Timmerman, Malvina Nissim

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

30 Downloads (Pure)

Abstract

As socially unacceptable language become pervasive in social media platforms, the need for automatic content moderation become more pressing. This contribution introduces the Dutch Abusive Language Corpus (DALC v1.0), a new dataset with tweets manually an- notated for abusive language. The resource ad- dress a gap in language resources for Dutch and adopts a multi-layer annotation scheme modeling the explicitness and the target of the abusive messages. Baselines experiments on all annotation layers have been conducted, achieving a macro F1 score of 0.748 for binary classification of the explicitness layer and .489 for target classification.
Original languageEnglish
Title of host publicationProceedings of the 5th Workshop on Online Abuse and Harm
EditorsAida Mostafazadeh Davani, Douwe Kiela, Mathias Lambert, Bertie Vidgen, Vinodkumar Prabhakaran, Zeerak Waseem
PublisherAssociation for Computational Linguistics (ACL)
Pages54-66
Number of pages13
DOIs
Publication statusPublished - 27-Jul-2021

Keywords

  • language models
  • hate speech
  • offensive language

Cite this