Elamur Research
☰
home / papers / 2017 / A Survey on Hate Speech Detection using Natural Language Processing

A Survey on Hate Speech Detection using Natural Language Processing

Written for the Research library (claude-opus-5-5) from the paper text. Original — aclanthology.org/W17-1101.pdf

Summary

This 2017 survey gives NLP researchers new to the field a structured map of automatic hate speech detection. The authors argue that simple keyword filtering is not enough, because whether a message is hateful depends on context, target, author and accompanying media. They first sort out the many overlapping labels in the literature, such as abusive messages, flames, cyberbullying, offensive language and othering language. They then group prior systems mainly by feature type: surface n-grams, word generalization (clusters, topics, embeddings), sentiment cues, lexicons, linguistic structure such as dependency relations, knowledge-based and meta-information, and multimodal signals. Most systems treat the task as supervised classification. Even general-purpose representations, like embeddings or bag-of-words vectors, already give fairly good results, and character-level models tend to beat token-level ones. Slur lists mostly help only when combined with other features. The authors stress that results are hard to compare because data sets are small, often private, limited to subtypes of hate speech, and annotated under vague guidelines. They call for a shared benchmark, a common definition, multilingual work, and better modelling of context.

Key points

  • Mehdad and Tetreault (2016) found character n-grams more predictive than token n-grams. Character models also cope better with obfuscated spellings like "ki11 yrslef a$$hole".
  • Word generalization approaches include Brown clustering, LDA topic distributions and word embeddings. According to Nobata et al. (2016), averaged word vectors work only modestly, while paragraph embeddings (Djuric et al., 2015) perform much better.
  • Sentiment is used either in multi-step pipelines (subjectivity, then polarity, then hate classifier) or as features. One example of the feature approach is the SentiStrength intensity scores used by Burnap et al. (2013).
  • Lexical resources include web lists (hatebase.org, noswearing.com, rsdb.org, Wikipedia slur lists) and task-specific lexicons. Examples of the latter are Razavi et al.'s Insulting and Abusing Language Dictionary and Gitari et al.'s hate verbs.
  • Annotation is a major weakness. Crowdsourced AMT labels diverge strongly from expert labels, and Ross et al. (2016) could not reach reliable agreement even with a definition. Besides English, the survey notes notable work only on Dutch and German.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Natural language processing

Cited by

404 papers of our library cite it (of the 140673 whose bibliographies we read); 12 of them have a page here, 392 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/w17-1101

@inproceedings{schmidt2017survey,
  title = {{A Survey on Hate Speech Detection using Natural Language Processing}},
  author = {Schmidt, Anna and Wiegand, Michael},
  booktitle = {Proceedings of the Fifth International Workshop on Natural Language Processing for Social Media},
  year = {2017},
  doi = {10.18653/v1/w17-1101},
  url = {https://doi.org/10.18653/v1/w17-1101}
}

References

35 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. David M. Blei, Andrew Ng, and Michael I. Jordan. 2003. Latent Dirichlet Allocation. Journal of Machine Learning Research, 3:993–1022. find our card
  2. Peter F. Brown, Peter V. deSouza, Robert L. Mercer, Vincent J. Della Pietra, and Jenifer C. Lai. 1992. Class-based n-gram models of natural language. Computational Linguistics, 18(4):467–479. find
  3. Nemanja Djuric, Jing Zhou, Robin Morris, Mihajlo Gr-bovic, Vladan Radosavljevic, and Narayan Bhamidi-pati. 2015. Hate speech detection with comment embeddings. In Proceedings of the 24th International Conference on World Wide Web, pages 29–30, New York, NY, USA. ACM. find
  4. P. Burnap and M. Williams. 2014. Hate speech, machine classification and statistical modelling of information flows on twitter: Interpretation and communication for policy decision making. In Internet, Policy and Politics Conference, Oxford, United Kingdom. find
  5. Njagi Dennis Gitari, Zhang Zuping, Hanyurwimfura Damien, and Jun Long. 2015. A lexicon-based approach for hate speech detection. International Journal of Multimedia and Ubiquitous Engineering, 10(4):215–230. find
  6. Pete Burnap and Matthew L. Williams. 2015. Cyber hate speech on twitter: An application of machine classification and statistical modeling for policy and decision making. Policy & Internet, 7(2):223–242. find
  7. Homa Hosseinmardi, Sabrina Arredondo Mattson, Rahat Ibn Rafiq, Richard Han, Qin Lv, and Shivakant Mishra. 2015. Detection of cyberbullying incidents on the instagram social network. CoRR, abs/1503.03909. — [arXiv:1503.03909](https://arxiv.org/abs/1503.03909) find
  8. Pete Burnap and Matthew L. Williams. 2016. Us and them: identifying cyber hate on twitter across multiple protected characteristics. EPJ Data Science, 5(1):1–15. find
  9. Irene Kwok and Yuzhou Wang. 2013. Locate the hate: Detecting tweets against blacks. In Marie desJardins and Michael L. Littman, editors, AAAI, pages 1621– 1622, Bellevue, Washington, USA. AAAI Press. find
  10. Pete Burnap, Omer F. Rana, Nick Avis, Matthew Williams, William Housley, Adam Edwards, Jeffrey Morgan, and Luke Sloan. 2013. Detecting tension in online communities with computational twitter analysis. Technological Forecasting and Social Change, pages 96–108, May. find
  11. Quoc Le and Tomas Mikolov. 2014. Distributed Representations of Sentences and Documents. In Proceedings of the International Conference on Machine Learning (JMLR), pages 1188–1196, Beijing, China. find our card
  12. Pete Burnap, Matthew L. Williams, Luke Sloan, Omer Rana, William Housley, Adam Edwards, Vincent Knight, Rob Procter, and Alex Voss. 2014. Tweeting the terror: modelling the social media reaction to the woolwich terrorist attack. Social Network Analysis and Mining, 4(1):1–14. find
  13. Hugo Liu and Push Singh. 2004. ConceptNet: A Practical Commonsense Reasoning Toolkit. BT Technology Journal, 22:211–226. find
  14. Yashar Mehdad and Joel Tetreault. 2016. Do characters abuse more than words? In 17th Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 299–303, Los Angeles, CA, USA. find
  15. Michael Chau and Jennifer Xu. 2007. Mining communities and their relationships in blogs: A study of online hate groups. International Journal of Human-Computer Studies, 65(1):57–70. find
  16. Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. 2013. Efficient Estimation of Word Representations in Vector Space. In Proceedings of Workshop at the International Conference on Learning Representations (ICLR), Scottsdale, AZ, USA. find
  17. Mike Thelwall, Kevan Buckley, Georgios Paltoglou, and Di Cai. 2010. Sentiment Strength Detection in Short Informal Text. Journal of the American Society for Information Science and Technology, 61(12):2544–2558. find
  18. Chikashi Nobata, Joel Tetreault, Achint Thomas, Yashar Mehdad, and Yi Chang. 2016. Abusive language detection in online user content. In Proceedings of the 25th International Conference on World Wide Web, pages 145–153, Geneva, Switzerland. find
  19. Cynthia Van Hee, Els Lefever, Ben Verhoeven, Julie Mennes, Bart Desmet, Guy De Pauw, Walter Daelemans, and Véronique Hoste. 2015. Detection and fine-grained classification of cyberbullying events. In Proceedings of Recent Advances in Natural Language Processing, Proceedings, pages 672–680, Hissar,… find
  20. John T. Nockleby. 2000. Hate Speech. In Leonard W. Levy, Kenneth L. Karst, and Dennis J. Mahoney, editors, Encyclopedia of the American Constitution, pages 1277–1279. Macmillan, 2nd edition. find
  21. Xiaofeng Wang, Matthew S Gerber, and Donald E Brown. 2012. Automatic crime prediction using events extracted from twitter posts. In International Conference on Social Computing, Behavioral-Cultural Modeling, and Prediction, pages 231–238. find
  22. Amir H. Razavi, Diana Inkpen, Sasha Uritsky, and Stan Matwin. 2010. Offensive language detection using multi-level classification. In Proceedings of the 23rd Canadian Conference on Advances in Artificial Intelligence, AI’10, pages 16–27, Berlin, Heidelberg. find
  23. William Warner and Julia Hirschberg. 2012. Detecting hate speech on the world wide web. In Proceedings of the Second Workshop on Language in Social Media, LSM ’12, pages 19–26, Stroudsburg, PA, USA. Association for Computational Linguistics. find
  24. Ellen Riloff, Ashequl Qadir, Prafulla Surve, Lalindra De Silva, Nathan Gilbert, and Ruihong Huang. 2013. Sarcasm as Contrast between a Positive Sentiment and Negative Situation. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 704–714, Seattle, WA,… find our card
  25. Zeerak Waseem and Dirk Hovy. 2016. Hateful symbols or hateful people? predictive features for hate speech detection on twitter. In Proceedings of the NAACL Student Research Workshop, pages 88–93, San Diego, California, USA, June. Association for Computational Linguistics. find
  26. Björn Ross, Michael Rist, Guillermo Carbonell, Benjamin Cabrera, Nils Kurowsky, and Michael Wojatzki. 2016. Measuring the Reliability of Hate Speech Annotations: The Case of the European Refugee Crisis. In Proceedings of the Workshop on Natural Language Processing for Computer-Mediated Communicatio… find
  27. Matthew Leighton Williams and Pete Burnap. 2015. Cyberhate on social media in the aftermath of woolwich: A case study in computational criminology and big data. British Journal of Criminology, pages 211–238. find
  28. Guang Xiang, Bin Fan, Ling Wang, Jason Hong, and Carolyn Rose. 2012. Detecting offensive tweets via topical feature discovery over a large scale twitter corpus. In Proceedings of the 21st ACM international conference on Information and knowledge management, pages 1980–1984, Maui, HI, USA. ACM. find
  29. Leandro Araújo Silva, Mainack Mondal, Denzil Correa, Fabrı́cio Benevenuto, and Ingmar Weber. 2016. Analyzing the targets of hate in online social media. In Proceedings of the Tenth International Conference on Web and Social Media, pages 687–690, Cologne, Germany. find
  30. Jun-Ming Xu, Kwang-Sung Jun, Xiaojin Zhu, and Amy Bellmore. 2012. Learning from bullying traces in social media. In Proceedings of the 2012 conference of the North American chapter of the association for computational linguistics: Human language technologies, pages 656–666, Montréal, Canada. Associ… find
  31. Sara Sood, Judd Antin, and Elizabeth Churchill. 2012a. Profanity use in online communities. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pages 1481–1490, Austin, TX, USA. ACM. find
  32. Haoti Zhong, Hao Li, Anna Cinzia Squicciarini, Sarah Michele Rajtmajer, Christopher Griffin, David J. Miller, and Cornelia Caragea. 2016. Content-driven detection of cyberbullying on the instagram social network. In IJCAI, pages 3952– 3958, New York City, NY, USA. IJCAI/AAAI Press. find
  33. Sara Owsley Sood, Elizabeth F. Churchill, and Judd Antin. 2012b. Automatic identification of personal insults on social news sites. J. Am. Soc. Inf. Sci. Technol., 63(2):270–285, February. find
  34. Ellen Spertus. 1997. Smokey: Automatic recognition of hostile messages. In Proceedings of the Fourteenth National Conference on Artificial Intelligence and Ninth Conference on Innovative Applications of Artificial Intelligence, AAAI’97/IAAI’97, pages 1058–1065, Providence, RI, USA. AAAI Press. find
  35. Yilu Zhou, Edna Reid, Jialun Qin, Hsinchun Chen, and Guanpi Lai. 2005. US Domestic Extremist Groups on the Web: Link and Content Analysis. IEEE intelligent systems, 20(5):44–51. find