Elamur Research⌕
☰
home / papers / 2019 / Challenges and frontiers in abusive content detection

Challenges and frontiers in abusive content detection

Summary

This position paper from The Alan Turing Institute examines automated detection of online abuse. Reported scores in this field have risen, but systems still struggle to generalize, run efficiently and meet real moderation needs. The authors do not propose a new model or dataset. Instead, they combine computational and social-science perspectives to map the remaining technical, social and ethical obstacles and to point to under-explored research directions.

A central argument is that tasks are often framed at the wrong level of abstraction. Some work targets vague categories such as "flagged" content, while other work narrows to "hate" and overlooks insults, dismissal or belittling. Drawing on Sartori's ladder of abstraction, the authors propose describing tasks along three dimensions: what the abuse targets, who receives it, and how it is expressed. They split directed abuse into individual-, identity- and concept-directed types, and they reject definitions based on speaker intent or presumed effects. The paper offers a conceptual framework for building clearer, more comparable datasets and tasks. It also gives guidance for researchers who work with harmful material.

Key points

  • Example of progress: Pitsilis et al. (2018) reached an F-score of 0.932 on Waseem and Hovy's dataset of about 16,000 tweets. The original 2016 result was 0.739.

  • Hidden dataset bias: The Davidson et al. (2017) dataset is widely used for "generic" abuse detection, yet it is heavily skewed toward racism and sexism.

  • Proposed typology of directed abuse:

    • individual-directed, split into tagged and referenced;
    • identity-directed, aimed at groups or affiliations;
    • concept-directed, aimed at ideologies, beliefs or countries.

    The authors caution against automatically moderating concept-directed abuse because of free-expression concerns.

  • Critique of terminology: The paper argues that definitions based on intent are unreliable. Examples include Pitsilis et al.'s definition of hate and Kumar et al.'s overt/covert distinction. Definitions based on effect are criticized as subjective, for example the "toxicity" label in Wulczyn et al.'s 100,000+ Wikipedia comments, which is tied to whether a comment makes people leave a discussion.

  • Venue: Third Workshop on Abusive Language Online (ACL), Florence, 2019, pp. 80–93.

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

106 papers of our library cite it (of the 140884 whose bibliographies we read); 10 of them have a page here, 96 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/w19-3509

@inproceedings{vidgen2019challenges,
  title = {{Challenges and frontiers in abusive content detection}},
  author = {Vidgen, Bertie and Harris, Alex and Nguyen, Dong and Tromble, Rebekah and Hale, Scott and Margetts, Helen},
  booktitle = {Proceedings of the Third Workshop on Abusive Language Online},
  year = {2019},
  doi = {10.18653/v1/w19-3509},
  url = {https://doi.org/10.18653/v1/w19-3509}
}

References

82 references, 13 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Ensuring that abusive content detection systems can be applied across different domains is one of the most difficult but also important frontiers in existing research. Thus far, efforts to address this has been unsuccessful. Burnap and Williams train systems on one type of hate speech (e.g. racism)… find
  2. Agrawal, S., & Awekar, A. (2018). Deep learning for detecting cyberbullying across multiple social media platforms. In ECIR: Advances in Information Retrieval (pp. 141–153). https://doi.org/10.1007/978-3-319-76941-7_11 — [doi:10.1007/978-3-319-76941-7_11](https://doi.org/10.1007/978-3-319-76941-7_11) ↗
  3. EPJ Data Science, 5(1), 1–15. https://doi.org/10.1140/epjds/s13688-016-0072-6 — [doi:10.1140/epjds/s13688-016-0072-6](https://doi.org/10.1140/epjds/s13688-016-0072-6) ↗
  4. Caiani, M., & Wagemann, C. (2009). Online networks of the Italian and German Extreme Right. Information, Communication & Society, 66–109. https://doi.org/10.1080/1369118080215848 2 — [doi:10.1080/1369118080215848](https://doi.org/10.1080/1369118080215848) ↗
  5. Allen, C. (2011). Islamophobia. Surrey: Ashgate. find
  6. Amichai-hamburger, Y., & McKenna, K. (2006). The Contact Hypothesis Reconsidered: Interacting via the Internet. Journal of Computer-Mediated Communication, 11(1), 825–843. https://doi.org/10.1111/j.1083-6101.2006.00037.x — [doi:10.1111/j.1083-6101.2006.00037.x](https://doi.org/10.1111/j.1083-6101.2006.00037.x) ↗
  7. Chandrasekharan, E., Samory, M., Srinivasan, A., & Gilbert, E. (2017). The Bag of Communities. In CHI (pp. 3175–3187). https://doi.org/10.1145/3025453.3026018 — [doi:10.1145/3025453.3026018](https://doi.org/10.1145/3025453.3026018) ↗
  8. Crawford, K., & Gillespie, T. (2016). What is a flag for? Social media reporting tools and the vocabulary of complaint. New Media and Society, 18(3), 410–428. https://doi.org/10.1177/1461444814543163 — [doi:10.1177/1461444814543163](https://doi.org/10.1177/1461444814543163) ↗
  9. Anzovino, M., Fersini, E., & Rosso, P. (2018). Automatic identification and classification of misogynistic language on Twitter. In NLDB (pp. 57–64). https://doi.org/10.1007/978-3-319-91947-8_6 — [doi:10.1007/978-3-319-91947-8_6](https://doi.org/10.1007/978-3-319-91947-8_6) ↗
  10. Daniels, J. (2013). Race and racism in Internet Studies: A review and critique. New Media and Society, 15(5), 695–719. https://doi.org/10.1177/1461444812462849 — [doi:10.1177/1461444812462849](https://doi.org/10.1177/1461444812462849) ↗
  11. Badjatiya, P., Gupta, M., & Varma, V. (2019). Stereotypical Bias Removal for Hate Speech Detection Task using Knowledge-based Generalizations. In World Wide Web (pp. 49–59). find
  12. Davidson, T., Warmsley, D., Macy, M., & Weber, I. (2017). Automated Hate Speech Detection and the Problem of Offensive Language. In ICWSM (pp. 1–4). find our card
  13. Badjatiya, P., Gupta, S., Gupta, M., & Varma, V. (2017). Deep Learning for Hate Speech Detection in Tweets. In World Wide Web (pp. 759–760). https://doi.org/10.1145/3041021.3054223 — [doi:10.1145/3041021.3054223](https://doi.org/10.1145/3041021.3054223) ↗
  14. Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. ArXiv:1810.04805v2, 1–16. Retrieved from http://arxiv.org/abs/1810.04805 ↗
  15. Benesch, S. (2012). Dangerous Speech: A Proposal to Prevent Group Violence. New York. find
  16. Dixon, L., Li, J., Sorensen, J., Thain, N., & Vasserman, L. (2018). Measuring and Mitigating Unintended Bias in Text Classification. In AAAI/ACM Conference on AI, Ethics, and Society (pp. 67–73). https://doi.org/10.1145/3278721.3278729 — [doi:10.1145/3278721.3278729](https://doi.org/10.1145/3278721.3278729) ↗
  17. Binns, R., Veale, M., Van Kleek, M., & Shadbolt, N. (2017). Like trainer, like bot? Inheritance of bias in algorithmic content moderation. In Lecture Notes in Computer Science (pp. 1–12). https://doi.org/10.1007/978-3-319-67256-4_32 — [doi:10.1007/978-3-319-67256-4_32](https://doi.org/10.1007/978-3-319-67256-4_32) ↗
  18. Eatwell, R. (2006). Community Cohesion and Cumulative Extremism in Contemporary Britain. Political Quarterly, 77(2), 204– 216. https://doi.org/10.1111/j.1467-923X.2006.00763.x — [doi:10.1111/j.1467-923x.2006.00763.x](https://doi.org/10.1111/j.1467-923x.2006.00763.x) ↗
  19. Buchanan, E. (2017). Considering the ethics of big data research: A case of Twitter and ISIS / ISIL. PLoS ONE, 12(12), 1–6. find
  20. Eger, S., Şahin, G. G., Rücklé, A., Lee, J.-U., Schulz, C., Mesgar, M., … Gurevych, I. (2019). Text Processing Like Humans Do: Visually Attacking and Shielding NLP Systems. ArXiv:1903.11508v1, 1–14. find
  21. Bucy, E., & Holbert, L. (2013). Sourcebook for political communication research. London: Routledge. find
  22. Burnap, P., & Williams, M. (2016). Us and Them: Identifying Cyber Hate on Twitter across Multiple Protected Characteristics. find
  23. Eisenstein, J. (2013). What to do about bad language on the Internet. NAACL HLT, 359–369. a European context. Critical Social Policy, 38(1), 5–12. https://doi.org/10.1177/0261018317732879 ↗ our card
  24. Eisenstein, J., O’Connor, B., Smith, N., & Xing, E. (2014). Diffusion of lexical change in social media. PLoS ONE, 9(11), 1–13. https://doi.org/10.1371/journal.pone.01131 14 — [doi:10.1371/journal.pone.01131](https://doi.org/10.1371/journal.pone.01131) ↗
  25. HM Government. (2019). Online Harms White Paper. London: Department of Digital, Culture, Media and Society. find
  26. Fortuna, P., & Nunes, S. (2018). A Survey on Automatic Detection of Hate Speech in Text. ACM Computing Surveys, 51(4), 1– 30. https://doi.org/10.1145/3232676 — [doi:10.1145/3232676](https://doi.org/10.1145/3232676) ↗
  27. Karan, M., & Šnajder, J. (2018). Cross-Domain Detection of Abusive Language Online. In 2nd Workshop on Abusive Language Online (pp. 132–137). Retrieved from http://takelab.fer.hr/alfeda ↗
  28. Founta, A., Djouvas, C., Chatzakou, D., Leontiadis, I., Blackburn, J., Stringhini, G., … Kourtellis, N. (2018). Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior. In ICWSM (pp. 1–11). find
  29. Kumar, R., Ojha, A. K., Malmasi, S., & Zampieri, M. (2018). Benchmarking Aggression Identification in Social Media. In 1st Workshop on Abusive Language Online (pp. 1–11). find our card
  30. Fraser, N. (1997). Justice Interruptus: Critical Reflections on the “Postsocialist” Condition. London: Routledge. find
  31. Kumar, R., Reganti, A., Bhatia, A., & Maheshwari, T. (2018). Aggression-annotated Corpus of Hindi-English Code-mixed Data. In LREC (pp. 1–7). Retrieved from http://arxiv.org/abs/1803.09402 ↗ our card
  32. Garg, N., Schiebinger, L., Jurafsky, D., & Zou, J. (2017). Word Embeddings Quantify 100 Years of Gender and Ethnic Stereotypes. PNAS, 115(16), 3635–3644. https://doi.org/10.1073/pnas.1720347115 — [doi:10.1073/pnas.1720347115](https://doi.org/10.1073/pnas.1720347115) ↗
  33. LaMarre, H., Landreville, K., & Beam, M. (2009). The irony of satire: Political ideology and the motivation to see what you want to see in The Colbert Report. International Journal of Press/Politics, 14(2), 212–231. https://doi.org/10.1177/1940161208330904 — [doi:10.1177/1940161208330904](https://doi.org/10.1177/1940161208330904) ↗
  34. Golbeck, J., Geller, A. A., Thanki, J., Naik, S., Hoffman, K. M., Wu, D. M., … Jienjitlert, V. (2017). A Large Labeled Corpus for Online Harassment Research. In WebSci (pp. 229–233). https://doi.org/10.1145/3091478.3091509 — [doi:10.1145/3091478.3091509](https://doi.org/10.1145/3091478.3091509) ↗
  35. Lee, Y., Yoon, S., & Jung, K. (2018). Comparative Studies of Detecting Abusive Language on Twitter. In 2nd Workshop on Abusive Language Online (pp. 101–106). find
  36. Gorrell, G., Greenwood, M., Roberts, I., Maynard, D., & Bontcheva, K. (2018). Twits, Twats and Twaddle: Trends in Online Abuse towards UK Politicians. In ICWSM (pp. 600–603). find
  37. Lipton, Z. (2016). The Mythos of Model Interpretability. In ICML Workshop on Human Interpretability in Machine Learning (pp. 1–9). New York. find
  38. Gröndahl, T., Pajola, L., Juuti, M., Conti, M., & Asokan, N. (2018). All You Need is “Love”: Evading Hate-speech Detection. ArXiv:1808.09115v2, 1–11. — [arXiv:1808.09115](https://arxiv.org/abs/1808.09115) find
  39. Ma, C. (2014). What are you laughing at? A social semiotic analysis of ironic racial stereotypes in Chappelle’s Show. London. find
  40. Hine, G. E., Onaolapo, J., De Cristofaro, E., Kourtellis, N., Leontiadis, I., Samaras, R., … Blackburn, J. (2017). Kek, Cucks, and God Emperor Trump: A Measurement Study of 4chan’s Politically Incorrect Forum and Its Effects on the Web. In ICWSM (pp. 92–101). find
  41. Magu, R., Joshi, K., & Luo, J. (2017). Detecting the Hate Code on Social Media. In ICWSM (pp. 608–611). https://doi.org/10.1016/j.vetimm.2017.02.0 03 — [doi:10.1016/j.vetimm.2017.02.0](https://doi.org/10.1016/j.vetimm.2017.02.0) ↗
  42. Margetts, H., John, P., Hale, S., & Yasseri, T. (2015). Political Turbluence: How Social Media Shape Collective Action. New Jersey: Princeton University Press. find
  43. Hines, S., Davy, Z., Monro, S., Motmans, J., Santos, A. C., & Van Der Ros, J. (2018). Introduction to the themed issue: Trans* policy, practice and lived experience within Marwick, A., Blackwell, L., & Lo, K. (2016). Best practices for conducting risky research and protecting yourself from online ha… find
  44. Tweets Using Deep Learning. ArXiv:1801.04433v1, 1–17. — [arXiv:1801.04433](https://arxiv.org/abs/1801.04433) find
  45. Qian, J., ElSherief, M., Belding, E., & Wang, W. Y. (2018). Hierarchical CVAE for Fine-Grained Hate Speech Classification. In EMNLP (pp. 3550–3559). https://doi.org/arXiv:1809.00088v1 find
  46. Marwick, A., & Boyd, D. (2010). I tweet honestly, I tweet passionately: Twitter users, context collapse, and the imagined audience. New Media and Society, 13(1), 114–133. https://doi.org/10.1177/1461444810365313 — [doi:10.1177/1461444810365313](https://doi.org/10.1177/1461444810365313) ↗
  47. Mehdad, Y., & Tetreault, J. (2016). Do Characters Abuse More Than Words? In SIGDAL (pp. 299–303). https://doi.org/10.18653/v1/w16-3638 ↗
  48. Raisi, E., & Huang, B. (2016). Cyberbullying Identification Using Participant-Vocabulary Consistency. In ICML Workshop on \#Data4Good (pp. 46–50). Retrieved from http://arxiv.org/abs/1606.08084 — [arXiv:1606.08084](https://arxiv.org/abs/1606.08084) ↗
  49. Müller, K., & Schwarz, C. (2017). Fanning the Flames of Hate: Social Media and Hate Crime. CAGE Working Paper Series, 1–82. https://doi.org/10.2139/ssrn.3082972 — [doi:10.2139/ssrn.3082972](https://doi.org/10.2139/ssrn.3082972) ↗
  50. Raisi, E., & Huang, B. (2017). Cyberbullying Detection with Weakly Supervised Machine Learning. In ASONAM (pp. 1–8). https://doi.org/10.1145/3110025.3110049 — [doi:10.1145/3110025.3110049](https://doi.org/10.1145/3110025.3110049) ↗
  51. Nobata, C., Thomas, A., Mehdad, Y., Chang, Y., & Tetreault, J. (2016). Abusive Language Detection in Online User Content. In World Wide Web (pp. 145–153). https://doi.org/10.1145/2872427.2883062 — [doi:10.1145/2872427.2883062](https://doi.org/10.1145/2872427.2883062) ↗
  52. Robinson, D., Zhang, Z., & Tepper, J. (2018). Hate speech detection on Twitter: Feature engineering v.s. feature selection. ESWC, 46–49. https://doi.org/10.1007/978-3-319-98192-5_9 — [doi:10.1007/978-3-319-98192-5_9](https://doi.org/10.1007/978-3-319-98192-5_9) ↗
  53. Palmer, A., Robinson, M., & Phillips, K. (2017). Illegal is not a Noun: Linguistic Form for Detection of Pejorative Nominalizations. In 1st Workshop on Abusive Language Online (pp. 91–100). https://doi.org/10.18653/v1/w17-3014 ↗
  54. Rudin, C. (2018). Please Stop Explaining Black Box Models for High Stakes Decisions. In NIPS (pp. 1–15). Retrieved from http://arxiv.org/abs/1811.10154 ↗ our card
  55. Ruppert, E., Law, J., & Savage, M. (2013). Reassembling Social Science Methods: The Challenge of Digital Devices. Theory, Culture & Society, 30(4), 22–46. https://doi.org/10.1177/0263276413484941 — [doi:10.1177/0263276413484941](https://doi.org/10.1177/0263276413484941) ↗
  56. Pan, S. J., & Fellow, Q. Y. (2009). A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, 1–15. find
  57. Papegnies, E., Labatut, V., Dufour, R., & Linarès, G. (2017). Graph-based features for automatic online abuse detection. In SLSP (pp. 70–81). https://doi.org/10.1007/978-3-319-68456-7_6 — [doi:10.1007/978-3-319-68456-7_6](https://doi.org/10.1007/978-3-319-68456-7_6) ↗
  58. Sabba, M. (2009). Spelling as a social practice. In J. Maybin & J. Swann (Eds.), Routledge Companion to English Language Studies (pp. 243–257). London: Routledge. find
  59. Sabou, M., Bontcheva, K., Derczynski, L., & Scharl, A. (2014). Corpus Annotation through Crowdsourcing: Towards Best Practice Guidelines. LREC, 859–866. find
  60. Park, Ji Ho, Shin, J., & Fung, P. (2018). Reducing Gender Bias in Abusive Language Detection. In EMNLP (pp. 2799– 2804). find our card
  61. Salminen, J., & Almerekhi, H. (2018). Online Hate Interpretation Varies by Country, But More by Individual. In SNAMS (pp. 1–7). https://doi.org/10.1109/SNAMS.2018.8554 954 — [doi:10.1109/snams.2018.8554](https://doi.org/10.1109/snams.2018.8554) ↗
  62. Park, Ji Hoon, Gabbadon, N. G., & Chernin, A. R. (2006). Naturalizing racial differences through comedy: Asian, black, and white views on racial stereotypes in rush hour 2. Journal of Communication, 56(1), 157– 177. https://doi.org/10.1111/j.1460-2466.2006.00008.x — [doi:10.1111/j.1460-2466.2006.00008.x](https://doi.org/10.1111/j.1460-2466.2006.00008.x) ↗
  63. Salminen, J., Almerekhi, H., Milenković, M., Jung, S.-G. G., An, J., Kwak, H., & Jansen, B. (2018). Anatomy of Online Hate: Developing a Taxonomy and Machine Learning Models. In ICWSM (pp. 330– 339). find
  64. Pitsilis, G. K., Ramampiaro, H., & Langseth, H. (2018). Detecting Offensive Language in https://doi.org/10.1109/SNAMS.2018.8554 954 https://blog.twitter.com/en_us/topics/comp any/2018/enabling-further-research-of-information-operations-on-twitter.html Sanguinetti, M., Poletto, F., Bosco, C., Patti,… ↗ our card
  65. Unsvåg, E., & Gambäck, B. (2018). The Effects of User Features on Twitter Hate Speech Detection. In 2nd Workshop on Abusive Language Online (pp. 75–85). Retrieved from http://aclweb.org/anthology/W18-5110 ↗
  66. Santos, C. N. dos, Melnyk, I., & Padhi, I. (2018). Fighting Offensive Language on Social Media with Unsupervised Text Style Transfer. In Meeting for ACL (pp. 189– 194). Retrieved from http://arxiv.org/abs/1805.07685 van Aken, B., Risch, J., Krestel, R., & Löser, A. (2018). Challenges for Toxic Comme… ↗ our card
  67. Sartori, G. (1970). Concept misinformation in comparative politics. American Political Science Review, 64(4), 1033–1053. https://doi.org/10.2307/1958356 — [doi:10.2307/1958356](https://doi.org/10.2307/1958356) ↗
  68. Van Hee, C., Lefever, E., De Pauw, G., Daelemans, W., Hoste, V., Jacobs, G., … Verhoeven, B. (2018). Automatic detection of cyberbullying in social media text. PLoS ONE, 13(10), 1–21. https://doi.org/10.1371/journal.pone.02037 94 ↗
  69. Schmidt, A., & Wiegand, M. (2017). A Survey on Hate Speech Detection using Natural Language Processing. In International Workshop on NLP for Social Media (pp. 1– 10). Valencia, Spain. https://doi.org/10.18653/v1/w17-1101 ↗ our card
  70. Waseem, Z., Davidson, T., Warmsley, D., & Weber, I. (2017). Understanding Abuse: A Typology of Abusive Language Detection Subtasks. In 1st Workshop on Abusive Language Online (pp. 78–84). https://doi.org/10.1080/1742177090311468 7 ↗ our card
  71. Serrà, J., Leontiadis, I., Spathis, D., Stringhini, G., Blackburn, J., & Vakali, A. (2017). Class-based Prediction Errors to Detect Hate Speech with Out-of-vocabulary Words. In 1st Workshop on Abusive Language Online (pp. 36–40). Vancouver, Canada. https://doi.org/10.18653/v1/w17-3005 ↗
  72. Waseem, Z., & Hovy, D. (2016). Hateful Symbols or Hateful People? Predictive Features for Hate Speech Detection on Twitter. In NAACL-HLT (pp. 88–93). https://doi.org/10.18653/v1/n16-2013 ↗ our card
  73. Silva, L., Mondal, M., Correa, D., Benevenuto, F., & Weber, I. (2016). Analyzing the Targets of Hate in Online Social Media. In IWCSM (pp. 687–690). find
  74. Weaver, S. (2010). Developing a rhetorical analysis of racist humour: examining anti-black jokes on the Internet. Social Semiotics, 20(5), 537–555. https://doi.org/10.1080/10350330.2010.513 188 — [doi:10.1080/10350330.2010.513](https://doi.org/10.1080/10350330.2010.513) ↗
  75. Sugandhi, R., Pande, A., Chawla, S., Agrawal, A., & Bhagat, H. (2016). Methods for detection of cyberbullying: a survey. In International Conference on Intelligent Systems Design and Applications (pp. 173– 177). https://doi.org/10.1109/ISDA.2015.748922 0 — [doi:10.1109/isda.2015.748922](https://doi.org/10.1109/isda.2015.748922) ↗
  76. Wiegand, M., Ruppenhofer, J., & Kleinbauer, T. (2019). Detection of Abusive Language: the Problem of Biased Datasets. In NAACL-HLT (pp. 602–608). find our card
  77. Tien, J. H., Eisenberg, M. C., Cherng, S. T., & Porter, M. A. (2019). Online reactions to the 2017 ‘Unite the Right’ rally in Charlottesville: Measuring polarization in Twitter networks using media followership. find
  78. Wulczyn, E., Thain, N., & Dixon, L. (2017). Ex Machina: Personal Attacks Seen at Scale. In World Wide Web (pp. 1391–1399). Perth, Australia. https://doi.org/10.1039/C5CC05843K — [doi:10.1039/c5cc05843k](https://doi.org/10.1039/c5cc05843k) ↗
  79. Twitter. (2018). Enabling further research of information operations on Twitter. Retrieved May 30, 2019, from Yao, M., Chelmis, C., & Zois, D. S. (2018). Cyberbullying detection on Instagram with optimal online feature selection. In ASONAM (pp. 401–408). https://doi.org/10.1109/ASONAM.2018.85 08329 — [doi:10.1109/asonam.2018.85](https://doi.org/10.1109/asonam.2018.85) ↗
  80. Zampieri, M., Malmasi, S., Nakov, P., Rosenthal, S., Farra, N., & Kumar, R. (2019). SemEval-2019 Task 6: Identifying and Categorizing Offensive Language in Social Media (OffensEval). ArXiv:1903.08983v3, 1–12. find our card
  81. Zannettou, S., Caulfield, T., Blackburn, J., & Cristofaro, E. De. (2018). On the Origins of Memes by Means of Fringe Web Communities. In 18th ACM Internet Measurement Conference (pp. 1–23). find
  82. Zhang, Z., Robinson, D., Tepper, J., Gangemi, A., Navigli, R., Vidal, M. E., … Alam, M. (2018). Detecting hate speech on Twitter using a convolution-GRU based deep neural network. In ESWC (pp. 745–760). https://doi.org/10.1007/978-3-319-93417-4_48 — [doi:10.1007/978-3-319-93417-4_48](https://doi.org/10.1007/978-3-319-93417-4_48) ↗

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/vidgen-2019-challenges-frontiers-abusive-content-detection/