CoChat

When the Timeline Meets the Pipeline: A Survey on Automated Cyberbullying Detection

Open access
Elsafoury, Fatma, Katsigiannis, Stamos, Pervez, Zeeshan, Ramzan, Naeem · 2021 · IEEE Access
Cited by 73

The DOI matches its Crossref record; 26 of the 26 references it cites were found in a registry or an index.

Abstract

Web 2.0 helped user-generated platforms to spread widely. Unfortunately, it also allowed for cyberbullying to spread. Cyberbullying has negative effects that could lead to cases of depression and low self-esteem. It has become crucial to develop tools for automated cyberbullying detection. The research on developing these tools has been growing over the last decade, especially with the recent advances in machine learning and natural language processing. Given the large body of work on this topic, it is vital to critically review the literature on cyberbullying within the context of these latest advances. In this paper, we survey the automated detection of cyberbullying. Our survey sheds light on some challenges and limitations for the field. The challenges range from defining cyberbullying, data collection, and feature representation to model selection, training, and evaluation. We also provide some suggestions for improving the task of cyberbullying detection. In addition to the survey, we propose to improve the task of cyberbullying detection by addressing some of the raised limitations: 1) Using recent contextual language models like BERT for the detection of cyberbullying; 2) Using slang-based word embeddings to generate better representations of the cyberbullying-related datasets. Our results show that BERT outperforms state-of-the-art cyberbullying detection models and deep learning models. The results also show that deep learning models initialized with slang-based word embeddings outperform deep learning models initialized with traditional word embeddings.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/access.2021.3098979

Cites

26 of 26 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Research works printed but not linked: 123

  1. 10.1007/s10579-020-09502-8 · DOI · Paper page
  2. 10.1109/ISDA.2013.6920758 · DOI · Paper page
  3. 10.1371/journal.pone.0243300 · DOI · Paper page
  4. emmery, 2019, Current limitations in cyberbullying detection: On evaluation criteria, reproducibility, and data scarcity, arXiv 1910 11922
  5. 10.1109/ACCESS.2019.2918354 · DOI · Paper page
  6. tarwani, 2017, Survey of cyberbulling detection on social media big-data, Int J Adv Res Comput Sci, 8, 831
  7. 10.1109/ICONIC.2018.8601278 · DOI
  8. 10.1109/ICMLA.2011.152 · DOI · Paper page
  9. 10.1016/j.avb.2015.05.012 · DOI
  10. 10.1109/EISIC.2017.43 · DOI
  11. 10.1109/EMS.2016.037 · DOI · Paper page
  12. 10.1109/TAFFC.2016.2531682 · DOI · Paper page
  13. 10.1109/ICICTM.2016.7890808 · DOI
  14. 10.1016/j.chb.2016.05.051 · DOI · Paper page
  15. belsey, 2005, Cyberbullying: An emerging threat to the ‘always on’ generation, Recuperado el, 5, 2010
  16. 10.1080/00220973.1993.9943832 · DOI · Paper page
  17. 10.1109/78.650093 · DOI · Paper page
  18. 2020, Natural Language Toolkit
  19. 10.1108/eb046814 · DOI · Paper page
  20. 10.1145/3041021.3054223 · DOI · Paper page
  21. 10.1201/9781315382159 · DOI · Paper page
  22. 10.1145/383952.383974 · DOI · Paper page
  23. 2020, Text Tokenization Utility Class
  24. 10.1093/jigpal/jzv048 · DOI · Paper page
  25. 10.1109/WI-IAT.2011.24 · DOI
  26. 10.1108/S2044-9968(2012)0000005016 · doi:10.1108/s2044-9968(2012)0000005016 · Paper page
  27. munezero, 2013, Antisocial behavior corpus for harmful language detection, Proc Federated Conf Comput Sci Inf Syst, 261
  28. 10.1109/IJCNN.2018.8489211 · DOI · Paper page
  29. 10.1109/ICCPCT.2014.7054943 · DOI
  30. 10.1109/FUZZ-IEEE.2018.8491557 · DOI · Paper page
  31. 10.1109/ICDMW.2014.170 · DOI · Paper page
  32. 10.1080/07421222.2019.1599500 · DOI · Paper page
  33. 10.1002/casp.2136 · DOI · Paper page
  34. 10.1080/02673843.2019.1669059 · DOI · Paper page
  35. 10.18653/v1/N16-2013 · DOI · Paper page
  36. 10.1145/3091478.3091487 · DOI · Paper page
  37. 10.1145/2833312.2849567 · DOI · Paper page
  38. kowalski, 2012, Cyberbullying bullying in the digital age
  39. 10.1371/journal.pone.0203794 · DOI · Paper page
  40. 10.1145/2808797.2809381 · DOI · Paper page
  41. 10.1145/2436256.2436274 · DOI · Paper page
  42. 10.4324/9780203818312 · DOI · Paper page
  43. raschka, 2014, Naive Bayes and text classification i–introduction and theory, arXiv 1410 5329
  44. 10.1016/j.ins.2016.04.019 · DOI · Paper page
  45. 10.1016/j.ins.2020.05.018 · DOI
  46. 10.1016/j.patrec.2017.10.014 · DOI
  47. 10.1016/j.knosys.2019.104945 · DOI
  48. smith, 2012, Cyberbullying and cyber aggression, Handbook of School Violence and School Safety, 111
  49. 10.24963/ijcai.2019/808 · DOI · Paper page
  50. 10.1201/9781315382159 · DOI · Paper page
  51. 10.1109/ICDMW.2017.22 · DOI
  52. rafiq, 2016, Investigating factors influencing the latency of cyberbullying detection, arXiv 1611 05419
  53. 10.1109/KST.2017.7886127 · DOI
  54. kansara, 2015, A framework for cyberbullying detection in social network, International Journal of Current Engineering and Technology, 5, 494
  55. 10.1145/3027063.3053169 · DOI · Paper page
  56. 10.1145/2872427.2883062 · DOI · Paper page
  57. tian, 2016, Investigating cyberbullying in social media: The case of Twitter, Proc KSU Conf Cybersecur Educ Res Pract
  58. mahmud, 2008, Detecting flames and insults in text, Proc Int Conf Natural Lang Process, 1
  59. 10.1145/2998181.2998279 · DOI · Paper page
  60. 10.1145/3274433 · DOI · Paper page
  61. krasnowska-kieras, 2019, A simple neural network for cyberbullying detection, Proc PolEval Workshop, 161
  62. 10.1145/2702123.2702534 · DOI · Paper page
  63. 10.1145/3331184.3331262 · DOI · Paper page
  64. wilson, 2020, Urban dictionary embeddings for slang NLP applications, Proc 12th Lang Resour Eval Conf, 4764
  65. radford, 2018, Improving language understanding by generative pre-training
  66. 10.1109/EIT.2018.8500110 · DOI
  67. 10.1145/2464464.2464499 · DOI · Paper page
  68. hinduja, 2014, Cyberbullying: Identification, Prevention and Response Cyberbullying Research Center
  69. 10.1109/DSW.2019.8755598 · DOI
  70. 10.1016/j.yapd.2013.04.004 · DOI · Paper page
  71. 10.1145/3366424.3382084 · DOI
  72. 10.1016/B978-0-12-809481-5.00014-6 · DOI
  73. 10.1109/ICACCI.2015.7275970 · DOI · Paper page
  74. waseem, 2018, Bridging the Gaps Multi Task Learning for Domain Transfer of Hate Speech Detection, 29
  75. 10.1016/j.sbspro.2014.10.259 · doi:10.1016/j.sbspro.2014.10.259
  76. pak, 2010, Twitter as a corpus for sentiment analysis and opinion mining, Proc 7th Int Conf Lang Resour Eval (LREC), 1
  77. 10.1097/01.NNR.0000313482.33917.7d · DOI · Paper page
  78. liu, 2010, Sentiment analysis and subjectivity, Handbook of Natural Language Processing, 2, 627
  79. 10.1016/j.ijcci.2016.07.002 · DOI
  80. 10.3115/1220575.1220619 · DOI · Paper page
  81. nguyen, 2017, Emo, love, and god: Making sense of urban dictionary, a crowd-sourced online dictionary, arXiv 1712 08647
  82. 10.1109/ICCONS.2018.8662942 · doi:10.1109/iccons.2018.8662942
  83. papasavva, 2020, Raiders of the lost kek: 3.5 years of augmented 4chan posts from the politically incorrect board, arXiv 2001 07487
  84. 10.1080/1369118X.2020.1726989 · DOI
  85. dang, 2020, Ensemble bert for classifying medication-mentioning tweets, Proc 5th Social Media Mining Health Appl Workshop Shared Task, 37
  86. nahar, 2013, Cyberbullying detection based on text-stream classification, Proc Conferences Res Pract Inf Technol Ser Austral Comput Soc, 146, 49
  87. 10.1109/ASONAM.2018.8508294 · DOI
  88. 10.1109/ICACCE.2018.8441728 · DOI
  89. mikolov, 2013, Linguistic regularities in continuous space word representations, Proc (NAACL-HLT), 746
  90. bretschneider, 2014, Detecting online harassment in social networks, Proc 35th Int Conf Inf Syst (ICIS), 1
  91. 10.1145/2743065.2743085 · DOI · Paper page
  92. tommasel, 2018, Features for detecting aggression in social media: An exploratory study, Proc 19th Simposio Argentino de Inteligencia Artif (ASAI)-JAIIO (CABA), 1
  93. 10.1145/3110025.3110049 · DOI · Paper page
  94. 10.1109/ASONAM.2018.8508240 · DOI · Paper page
  95. 10.1007/978-3-319-13647-9_21 · DOI
  96. 10.1007/978-3-319-93417-4_48 · DOI · Paper page
  97. 10.1007/978-3-319-27433-1_4 · DOI · Paper page
  98. 10.1145/3308560.3316505 · DOI
  99. sutinen, 2014, Automatic detection of antisocial behaviour in texts, Informatica (Ljubljana), 38, 1
  100. 10.1145/3289600.3291037 · DOI · Paper page
  101. 10.1109/ASONAM.2016.7752342 · DOI · Paper page
  102. 10.1142/S0219622002000178 · doi:10.1142/s0219622002000178
  103. 10.1007/978-3-319-06483-3_25 · DOI
  104. 10.18653/v1/W16-5618 · DOI · Paper page
  105. 10.1007/978-981-13-2354-6_24 · DOI
  106. xu, 2012, Learning from bullying traces in social media, Proc Conf North Amer Chapter Assoc Comput Linguistics Hum Lang Technol, 656
  107. nalinipriya, 2015, A dynamic cognitive system for automatic detection and prevention of cyber-bullying attacks, ARPN J Eng Appl Sci, 10, 4618
  108. dinakar, 2011, Modeling the detection of textual cyberbullying, Proc Social Mobile Web Papers (ICWSM) Workshop, 1
  109. 10.1007/978-3-319-71249-9_4 · DOI
  110. koufakou, 2020, Florunito@trac-2: Retrofitting word embeddings on an abusive lexicon for aggressive language detection, Proc 2nd Workshop Trolling Aggression Cyberbullying (TRAC LREC), 106
  111. wang, 2020, A comparative study on word embeddings in deep learning for text classification, Proc 4th Int Conf Natural Lang Process Inf Retr, 37
  112. voué, 2020, 4chan & 8chan embeddings, arXiv 2005 06946
  113. silva, 2016, Analyzing the targets of hate in online social media, Proc Int AAAI Conf Web Social Media, 10, 1
  114. wilson, 2020, Urban dictionary embeddings for slang NLP applications, Proc 12th Lang Resour Eval Conf (LREC), 4764
  115. 10.1109/ASONAM.2018.8508329 · DOI
  116. 10.1145/3342220.3344933 · DOI
  117. 10.1016/j.procs.2015.03.085 · DOI · Paper page
  118. 10.1109/NCG.2018.8593146 · DOI · Paper page
  119. 10.1016/j.ins.2020.05.084 · DOI
  120. huang, 2018, Cyberbullying intervention based on convolutional neural networks, Proc 1st Workshop Trolling Aggression Cyberbullying (TRAC), 42
  121. 10.18653/v1/S19-2007 · DOI · Paper page
  122. 10.1109/BDCloud.2014.87 · DOI · Paper page
  123. 10.1145/3167132.3167317 · DOI

Cited by

See all 73

Metadata

Authors
Elsafoury, F., Katsigiannis, S., Pervez, Z., & Ramzan, N.
Venue
IEEE Access
Published
2021
Volume
9
Pages
103541-103563
ISSN
2169-3536 (ISSN-L)
OpenAlex
W3186778784
Cited by
73
References
26
FWCI
4.32
Type
journal article
Language
en
Open access
gold
Funders
UK Research and Innovation (UKRI) as a Knowledge Transfer Partnership (11465)

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.