Implicitly Abusive Language – What does it actually look like and why are we not getting there?

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 576–587 June 6–11, 2021. ©2021 Association for Computational Linguistics 576 Implicitly Abusive Language – What does it actually look like and why are we not getting there? Michael Wiegand Digital Age Research Center (D!ARC) Alpen-Adria-Universität Klagenfurt AT-9020 Klagenfurt, Austria michael.wiegand@aau.at Josef Ruppenhofer Leibniz Institute for German Language D-68161 Mannheim, Germany ruppenhofer@ids-mannheim.de Elisabeth Eder Institut für Germanistik Alpen-Adria-Universität Klagenfurt AT-9020 Klagenfurt, Austria elisabeth.eder@aau.at Abstract Abusive language detection is an emerging field in natural language processing which has received a large amount of attention recently. Still the success of automatic detection is lim- ited. Particularly, the detection of implicitly abusive language, i.e. abusive language that is not conveyed by abusive words (e.g. dumbass or scum), is not working well. In this posi- tion paper, we explain why existing datasets make learning implicit abuse difficult and what needs to be changed in the design of such datasets. Arguing for a divide-and-conquer strategy, we present a list of subtypes of implic- itly abusive language and formulate research tasks and questions for future research. 1 Introduction Abusive or offensive language is commonly de- fined as hurtful, derogatory or obscene utterances made by one person to another person or group of persons.1 Examples are (1)-(3). In the literature, closely related terms include hate speech (Waseem and Hovy, 2016) or cyberbullying (Zhong et al., 2016). While there may be nuanced differences in meaning, they are all compatible with the general definition above.2 (1) stop editing this, you dumbass. (2) Just want to slap the stupid out of these bimbos!!! (3) Go lick a pig you arab muslim piece of scum. Due to the rise of user-generated web content, the amount of abusive language is growing. NLP methods are required to focus human review efforts towards the most relevant microposts. 1http://thelawdictionary.org/abusive-language 2The examples in this work are included to illustrate the severity of abusive language. They are taken from actual web data and in no way reflect the opinion of the authors. Though there has been much work on abusive language detection in general, comparatively little work has been focusing on implicit forms of abu- sive language (4)-(5) (Waseem et al., 2017). By implicit abuse we understand abusive language that is not conveyed by (unambiguously) abusive words (e.g. dumbass, bimbo, scum). (4) I haven’t had an intelligent conversation with a woman in my whole life. (5) Why aren’t there any Mexicans on Star Trek? Because they don’t work in the future either. Detailed analyses of the output of existing classi- fiers have also revealed that currently only explicit abuse can be reliably detected (van Aken et al., 2018; Wiegand et al., 2019). In this position paper, we want to shed more light on the nature of implicitly abusive language. We identify subtypes of implicit abuse that can be found in existing datasets and the literature. We also outline shortcomings that prevent implicitly abusive language from really being learned …

Content truncated to 500 words