An Annotated Corpus for Sexism Detection in French Tweets

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 1397–1403 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 1397 An Annotated Corpus for Sexism Detection in French Tweets Patricia Chiril1, Véronique Moriceau1, Farah Benamara1 Alda Mari2, Gloria Origgi2, Marlène Coulomb-Gully3 (1) IRIT, Université de Toulouse, Université Toulouse III - UPS, France {firstname.lastname}@irit.fr (2) Institut Jean Nicod, CNRS, ENS, EHESS, Paris, France {firstname.lastname}@ens.fr (3) LERASS, Université de Toulouse, UT2J, France Abstract Social media networks have become a space where users are free to relate their opinions and sentiments which may lead to a large spreading of hatred or abusive messages which have to be moderated. This paper presents the first French corpus annotated for sexism detection composed of about 12,000 tweets. In a context of offensive content mediation on social media now regulated by European laws, we think that it is important to be able to detect automatically not only sexist content but also to identify if a message with a sexist content is really sexist (i.e. addressed to a woman or describing a woman or women in general) or is a story of sexism experienced by a woman. This point is the novelty of our annotation scheme. We also propose some preliminary results for sexism detection obtained with a deep learning approach. Our experiments show encouraging results. Keywords: sexism detection, social media, corpus, speech acts 1. Introduction Social media networks such as Facebook, Twitter, blogs and forums, have become a space where users are free to relate events, personal experiences, but also opinions and sentiments about products, events or other people. This may lead to a large spreading of hatred or abusive messages which have to be moderated. In particular, theses messages may express threats, harassment, intimidation or "disparage a person or a group on the basis of some characteristic such as race, color, ethnicity, gender, sexual orientation, nation- ality, religion, or other characteristics" (Nockleby, 2000). Although some countries, such as the United States, where hate speech is protected under the First Amendment as free- dom of expression (Massaro, 1990), many other countries, such as France, have laws prohibiting it, laws that extend to the internet and social media. For instance, since the French law of 27 January 2017 related to equality and citizenship, penalties due to discrimination are doubled and sexism is now considered as an aggravating factor. Sexism is prejudice or discrimination based on a person’s gender. It is based on the belief that one sex or gender is superior to another and it mainly affects women and girls. It can take several forms: sexist remarks, gestures, behaviours, practices, from insults to rape or murder. As mentioned in Combating Sexist Hate Speech, a report of the Council of Europe1, "the aim of sexist hate speech is to humiliate or objectify women, to undervalue their skills and opinions, to destroy their reputation, to make them feel vulnerable and fearful, and to control and punish them for not following a certain …

Content truncated to 500 words