Improving Hate Speech Type and Target Detection with Hateful Metaphor Features

We study the usefulness of hateful metaphorsas features for the identification of the type and target of hate speech in Dutch Facebook comments. For this purpose, all hateful metaphors in the Dutch LiLaH corpus were annotated and interpreted in line with Conceptual Metaphor Theory and Critical Metaphor Analysis. We provide SVM and BERT/RoBERTa results, and investigate the effect of different metaphor information encoding methods on hate speech type and target detection accuracy. The results of the conducted experiments show that hateful metaphor features improve model performance for the both tasks. To our knowledge, it is the first time that the effectiveness of hateful metaphors as an information source for hatespeech classification is investigated.

[1]  Nikola Ljubesic,et al.  The LiLaH Emotion Lexicon of Croatian, Dutch and Slovene , 2020, PEOPLES.

[2]  Kevin W. Saunders.  What about Hate Speech , 2011 .

[3]  Ona de Gibert,et al.  Hate Speech Dataset from a White Supremacy Forum , 2018, ALW.

[4]  Preslav Nakov,et al.  Predicting the Type and Target of Offensive Posts in Social Media , 2019, NAACL.

[5]  Beata Beigman Klebanov,et al.  A Report on the 2020 VUA and TOEFL Metaphor Detection Shared Task , 2020, FIGLANG.

[6]  Haizhou Li,et al.  Named-Entity Tagging and Domain adaptation for Better Customized Translation , 2018, NEWS@ACL.

[7]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[8]  Bettina Berendt,et al.  RobBERT: a Dutch RoBERTa-based Language Model , 2020, FINDINGS.

[9]  Ming-Wei Chang,et al.  BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding , 2019, NAACL.

[10]  Nazli Goharian,et al.  Hate speech detection: Challenges and solutions , 2019, PloS one.

[11]  Walter Daelemans,et al.  Exploring Stylometric and Emotion-Based Features for Multilingual Cross-Domain Hate Speech Detection , 2021, WASSA.

[12]  Preslav Nakov,et al.  SemEval-2020 Task 12: Multilingual Offensive Language Identification in Social Media (OffensEval 2020) , 2020, SEMEVAL.

[13]  Tommaso Caselli,et al.  I Feel Offended, Don’t Be Abusive! Implicit/Explicit Messages in Offensive and Abusive Language , 2020, LREC.

[14]  J. Charteris-Black.  Corpus Approaches to Critical Metaphor Analysis , 2004 .

[15]  Fumiyo Fukumoto,et al.  DeepMet: A Reading Comprehension Paradigm for Token-level Metaphor Detection , 2020, FIGLANG.

[16]  Ingmar Weber,et al.  Automated Hate Speech Detection and the Problem of Offensive Language , 2017, ICWSM.

[17]  G. Lakoff,et al.  Metaphors We Live by , 1982 .

[18]  Preslav Nakov,et al.  SemEval-2019 Task 6: Identifying and Categorizing Offensive Language in Social Media (OffensEval) , 2019, *SEMEVAL.

[19]  Tommaso Caselli,et al.  BERTje: A Dutch BERT Model , 2019, ArXiv.

[20]  Lysandre Debut,et al.  HuggingFace's Transformers: State-of-the-art Natural Language Processing , 2019, ArXiv.

[21]  Omer Levy,et al.  RoBERTa: A Robustly Optimized BERT Pretraining Approach , 2019, ArXiv.

[22]  Inga Dervinyt.  Metaphors in the Language of the Press : a Contrastive Analysis , 2010 .

[23]  Tomaz Erjavec,et al.  The FRENK Datasets of Socially Unacceptable Discourse in Slovene and English , 2019, TSD.

[24]  Beata Beigman Klebanov,et al.  A Report on the 2018 VUA Metaphor Detection Shared Task , 2018, Fig-Lang@NAACL-HLT.

[25]  Ralf Krestel,et al.  Challenges for Toxic Comment Classification: An In-Depth Error Analysis , 2018, ALW.

[26]  Ritesh Kumar,et al.  Benchmarking Aggression Identification in Social Media , 2018, TRAC@COLING 2018.

[27]  Lucia Specia,et al.  Guiding Neural Machine Translation Decoding with External Knowledge , 2017, WMT.