LIIR at SemEval-2020 Task 12: A Cross-Lingual Augmentation Approach for Multilingual Offensive Language Identification

Proceedings of the 14th International Workshop on Semantic Evaluation, pages 2073–2079 Barcelona, Spain (Online), December 12, 2020. 2073 LIIR at SemEval-2020 Task 12: A Cross-Lingual Augmentation Approach for Multilingual Offensive Language Identification Erfan Ghadery Department of Computer Science KU Leuven erfan.ghadery@student.kuleuven.be Marie-Francine Moens Department of Computer Science KU Leuven Sien.Moens@cs.kuleuven.be Abstract This paper presents our system entitled ‘LIIR’ for SemEval-2020 Task 12 on Multilingual Offen- sive Language Identification in Social Media (OffensEval 2). We have participated in Subtask A for English, Danish, Greek, Arabic, and Turkish languages. We adapt and fine-tune the BERT and multilingual Bert models made available by Google AI1 for English and non-English languages respectively. For the English language, we use a combination of two fine-tuned BERT models. For other languages, we propose a cross-lingual augmentation approach in order to enrich train- ing data and we use multilingual BERT to obtain sentence representations. LIIR achieved rank 14/38, 18/47, 24/86, 24/54, and 25/40 in Greek, Turkish, English, Arabic, and Danish languages, respectively. 1 Introduction Nowadays, with an exponential increase in the use of social media platforms such as Facebook and Twitter by people from different educational and cultural backgrounds, the need for automatic methods for recognizing and filtering offensive languages is necessary (Chen et al., 2012; Nobata et al., 2016). Different types of offensive content like hate speech (Malmasi and Zampieri, 2018), aggression (Kumar et al., 2018) and cyberbullying (Dinakar et al., 2011) can be very harmful to the user’s mental health, especially to children and youth (Xu and Zhu, 2010). The OffensEval 2019 competition (Zampieri et al., 2019b) was an attempt to build systems capable of recognizing offensive content in social networks for the English language. The OffensEval 2019 organizers defined three Subtasks: whether a message is offensive or not (Subtask A), what is the type of the offensive message (Subtask B), and who is the target of the offensive message (Subtask C). This year, they have extended the competition to several languages while the Subtasks remain the same as in OffensEval 2019. OffensEval 2020 (Zampieri et al., 2020) features a multilingual dataset with five languages including English, Danish, Turkish, Greek, and Arabic. This article presents our approaches to SemEval-2020 Task 12: OffensEval 2 - Multilingual Offensive Language Identification in Social Media. We have participated in Subtask A for all languages. The goal of Subtask A is recognizing if a sentence is offensive or not. For the English language, we separately fine- tune two bidirectional encoder representations of the BERT transformer architecture (Devlin et al., 2018) on two different datasets, and we use the combination of these two models for training our classifier. Also, we perform an extensive preprocessing for the English language. For other languages, we enhance the provided training dataset for each language using a cross-lingual augmentation approach, then, we train a classifier by fine-tuning a multilingual BERT (mBERT) (Devlin et al., 2018) using the augmented dataset with a linear classification layer on top. Our proposed augmentation approach, inspired by the works of (Lample and …

Content truncated to 500 words