CyberAgressionAdo-v1: a Dataset of Annotated Online Aggressions in French Collected through a Role-playing Game

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 867–875 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 867 CyberAgressionAdo-v1: a Dataset of Annotated Online Aggressions in French Collected through a Role-playing Game Ana¨ıs Ollagnier1, Elena Cabrio1, Serena Villata1, Catherine Blaya2 1Universit´e Cˆote d’Azur, Inria, CNRS, I3S 930 route des Colles, BP 145, 06903 Sophia Antipolis Cedex, France 2Universit´e Cˆote d’Azur, CNRS, Unite de Recherche Migrations et Societe (Urmis) Pˆole Universitaire Saint-Jean d’Angely, 24 avenue des Diables Bleus, F-06357 Nice Cedex 4 {anais.ollagnier,elena.cabrio,serena.villata,catherine.blaya}@univ-cotedazur.fr Abstract Over the past decades, the number of episodes of cyber aggression occurring online has grown substantially, especially among teens. Most solutions investigated by the NLP community to curb such online abusive behaviors consist of supervised approaches relying on annotated data extracted from social media. However, recent studies have highlighted that private instant messaging platforms are major mediums of cyber aggression among teens. As such interactions remain invisible due to the app privacy policies, very few datasets collecting aggressive conversations are available for the computational analysis of language. In order to overcome this limitation, in this paper we present the CyberAgressionAdo-V1 dataset, containing aggressive multiparty chats in French collected through a role-playing game in high-schools, and annotated at different layers. We describe the data collection and annotation phases, carried out in the context of a EU and a national research projects, and provide insightful analysis on the different types of aggression and verbal abuse depending on the targeted victims (individuals or communities) emerging from the collected data. Keywords: hate speech, cyber aggression, role-playing game 1. Introduction The Web and social media platforms provide an attrac- tive environment to young people to communicate with their peers, to establish social ties, and bring new op- portunities for learning (Scheidt, 2015). At the same time, such environment is also raising concerns about the ethical use of technology and expose people to counterproductive and unsafe interactions that set their mental health and well-being at high risk. Cyber ag- gressions are defined by (Grigg, 2010) as “intentional harm delivered by the use of electronic means to a per- son or a group of people irrespective of their age, who perceive(s) such acts as offensive, derogatory, harm- ful, or unwante” (p. 152). Aggressive messages may be published both as public posts on social media, but may also be sent in an insidious way on private instant messaging platforms (Sprugnoli et al., 2018). The pos- sibilities offered by social networks to share privately content among users combined with the increasing dig- ital literacy of teenagers has the paradoxical effect to hinder the possibility to study the actual cyberhate ac- tivities (Aizenkot and Kashy-Rosenbaum, 2018). Within the Natural Language Processing (NLP) com- munity, in the past few years there have been several efforts made to deal with the problem of online hate speech detection, leading to the creation of a number of datasets for hate speech detection in different lan- guages, mainly containing messages publicly posted …

Content truncated to 500 words