Leveraging Advanced NLP Techniques and Data Augmentation to Enhance Online Misogyny Detection
Source abstract
Abstract
Online misogyny is a significant societal challenge that reinforces gender inequalities and discourages women from engaging fully in digital spaces. Traditional moderation methods often fail to address the dynamic and context-dependent nature of misogynistic language, making adaptive solutions essential. This study presents a framework that integrates advanced natural-language processing techniques with strategic data augmentation to improve the detection of misogynistic content. Key contributions include emoji decoding to interpret symbolic communication, contextual expansion using Sentence-Transformer models, and LDA-based topic modeling to enhance data richness and contextual understanding. The framework incorporates machine-learning, deep-learning, and Transformer-based models to handle complex and nuanced language. Performance analysis highlights the effectiveness of the selected models, and comparative results emphasize the transformative role of data augmentation. This augmentation significantly enhanced model robustness, improved generalization, and strengthened the detection of misogynistic content.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Hate Speech and Cyberbullying Detection · Advanced Malware Detection Techniques · Cybercrime and Law Enforcement Studies
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Computing, University of Portsmouth, Portsmouth PO1 2UP, UK
School of Education, Languages and Linguistics, University of Portsmouth, Portsmouth PO1 2UP, UK
Funding
Faculty of Technology at the University of Portsmouth
References (45)
Show source-provided references
1. Special issue on online misogyny10.1080/14680777.2018.1447345 ↗
2. Poland, B. (2016). Haters: Harassment, Abuse, and Violence Online, University of Nebraska Press.10.2307/j.ctt1fq9wdp ↗
3. Step Together (2024, November 11). Mental Health and Violent Misogyny Online. n.d, Available online: https://steptogether.nsw.gov.au/violent-extremism/mental-health-and-violent-misogyny-online.html.
4. Tech Safety Australia (2024, November 11). Online Misogyny and Abuse. Available online: https://techsafety.org.au/wp-content/uploads/2019/11/HDT_TSA_OnlineMisogynyAndAbuse_V1.0.pdf.
5. “I don’t hate all women, just those stuck-up bitches”: How incels and mainstream pornography speak the same extreme language of misogyny10.1177/1077801221996453 ↗
6. Tranchese, A. (2024). “Like shagging a dead fish”: Misogyny and consent in online sex buyers’ reviews. Fem. Media Stud., 1–20.10.1080/14680777.2024.2431599 ↗
7. Reconsidering self-moderation: The role of research in supporting community-based models for online content moderation10.1145/3415178 ↗
8. Social media cyberbullying detection using machine learning
9. Amer, E., Kwak, K.S., and El-Sappagh, S. (2022). Context-based fake news detection model relying on deep learning models. Electronics, 11.10.3390/electronics11081255 ↗
10. Nayel, H., Amer, E., Allam, A., and Abdallah, H. (2021, January 19). Machine learning-based model for sentiment and sarcasm detection. Proceedings of the Sixth Arabic Natural Language Processing Workshop, Kyiv, Ukraine.
11. Boitel, E., Mohasseb, A., and Haig, E. (2023, January 6–8). A comparative analysis of GPT-3 and BERT models for text-based emotion recognition: Performance, efficiency, and robustness. Proceedings of the UK Workshop on Computational Intelligence, Birmingham, UK.10.1007/978-3-031-47508-5_44 ↗
12. Canós, J.S. (2018, January 18). Misogyny Identification Through SVM at IberEval 2018. Proceedings of the IberEval@SEPLN, Seville, Spain.
13. Liu, H., Chiroma, F., and Cocea, M. (2018, January 18). Identification and classification of misogynous tweets using multi-classifier fusion. Proceedings of the Evaluation of Human Language Technologies for Iberian Languages: IberEval 2018. CEUR Workshop Proceedings, Seville, Spain.
14. Gender bias perpetuation and mitigation in AI technologies: Challenges and opportunities10.1007/s00146-023-01675-4 ↗
15. Misogyny detection in twitter: A multilingual and cross-domain study10.1016/j.ipm.2020.102360 ↗
16. Harnessing the power of llms in practice: A survey on chatgpt and beyond
17. Recent advancements and challenges of nlp-based sentiment analysis: A state-of-the-art review10.1016/j.nlp.2024.100059 ↗
18. An emoji feature-incorporated multi-view deep learning for explainable sentiment classification of social media reviews10.1016/j.techfore.2024.123326 ↗
19. A context-aware attention and graph neural network-based multimodal framework for misogyny detection10.1016/j.ipm.2024.103895 ↗
20. Categorizing Sexism and Misogyny through Neural Approaches10.1145/3457189 ↗
21. Lynn, T., Endo, P.T., Rosati, P., Silva, I., Santos, G.L., and Ging, D. (2021, January 3–10). A Comparison of Machine Learning Approaches for Detecting Misogynistic Speech in Urban Dictionary. Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, Virtual.
22. Vetagiri, A., Mogha, P., and Pakray, P. (2024, January 9–12). Cracking Down on Digital Misogyny with MULTILATE: A MULTImodaL hATE Detection System. Proceedings of the Conference and Labs of the Evaluation Forum (CLEF), Grenoble, France.
23. Automatic Misogyny Detection in Social Media: A Survey10.13053/cys-23-4-3299 ↗
24. Online Hate Speech against Women: Automatic Identification of Misogyny and Sexism on Twitter10.3233/JIFS-179023 ↗
25. Li, K. (2022, January 26–28). An Evaluation of Automation on Misogyny Identification(AMI) and Deep-Learning Approaches for Hate Speech -Highlight on Graph Convolutional Networks and Neural Networks. Proceedings of the 2022 International Conference on Computers, Information Processing and Advanced Education (CIPAE), Ottawa, ON, Canada.10.1109/CIPAE55637.2022.00058 ↗
26. Rahali, A., Akhloufi, M.A., Therien-Daniel, A.M., and Brassard-Gourdeau, E. (2021, January 17–20). Automatic Misogyny Detection in Social Media Platforms using Attention-based Bidirectional-LSTM. Proceedings of the 2021 IEEE International Conference on Systems, Man, and Cybernetics (SMC), Melbourne, Australia.10.1109/SMC52423.2021.9659158 ↗
27. Anzovino, M., Fersini, E., and Rosso, P. (2018, January 13–15). Automatic Identification and Classification of Misogynistic Language on Twitter. Proceedings of the 2018 International Conference on Natural Language & Information Systems, Paris, France.10.1007/978-3-319-91947-8_6 ↗
28. Mienye, I.D., Swart, T.G., and Obaido, G. (2024). Recurrent neural networks: A comprehensive review of architectures, variants, and applications. Information, 15.10.20944/preprints202408.0748.v1 ↗
29. Safi Samghabadi, N., Patwa, P., PYKL, S., Mukherjee, P., Das, A., and Solorio, T. (2020, January 11–16). Aggression and Misogyny Detection using BERT: A Multi-Task Approach. Proceedings of the Second Workshop on Trolling, Aggression and Cyberbullying. Language Resources and Evaluation Conference (LREC), Marseille, France.
30. Morbidoni, C., and Sarra, A. (2023, January 20–22). Can LLMs assist humans in assessing online misogyny? Experiments with GPT-3.5. Proceedings of the GENerative, Explainable and Reasonable Artificial Learning Workshop 2023. In Conjunction with CHITALY 2023, Turin, Italy.
31. Youssif, A.A., Ghalwash, A.Z., and Amer, E.A. (2011, January 21–23). Hsws: Enhancing efficiency of web search engine via semantic web. Proceedings of the International Conference on Management of Emergent Digital EcoSystems, San Francisco, CA, USA.10.1145/2077489.2077530 ↗
32. Latent Dirichlet allocation (LDA) and topic modeling: Models, applications, a survey10.1007/s11042-018-6894-4 ↗
33. Comic exploration and Insights: Recent trends in LDA-Based recognition studies10.1016/j.eswa.2024.124732 ↗
34. A Semantics-enhanced Topic Modelling Technique: Semantic-LDA10.1145/3639409 ↗
35. Introduction to convolutional neural networks
36. Graves, A., and Graves, A. (2012). Long short-term memory. Supervised Sequence Labelling with Recurrent Neural Networks, Springer.10.1007/978-3-642-24797-2 ↗
37. Bidirectional LSTM with attention mechanism and convolutional layer for text classification10.1016/j.neucom.2019.01.078 ↗
38. Devlin, J., Chang, M.W., Lee, K., and Toutanova, K. (2018). Bert: Bidirectional encoder representations from transformers. arXiv.
39. Liu, Y. (2019). Roberta: A robustly optimized bert pretraining approach. arXiv, 364.
40. Sanh, V., Debut, L., Chaumond, J., and Wolf, T. (2019). DistilBERT, a distilled version of BERT: Smaller, faster, cheaper and lighter. arXiv.
41. Yang, Z. (2019). XLNet: Generalized Autoregressive Pretraining for Language Understanding. arXiv.
42. Lan, Z. (2019). Albert: A lite bert for self-supervised learning of language representations. arXiv.
43. Clark, K. (2020). Electra: Pre-training text encoders as discriminators rather than generators. arXiv.
44. Guest, E., Vidgen, B., Mittos, A., Sastry, N., Tyson, G., and Margetts, H. (2021, January 19–23). An Expert Annotated Dataset for the Detection of Online Misogyny. Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, Kyiv, Ukraine.10.18653/v1/2021.eacl-main.114 ↗
45. Yacouby, R., and Axman, D. (2020, January 20). Probabilistic extension of precision, recall, and F1 score for more thorough evaluation of classification models. Proceedings of the First Workshop on Evaluation and Comparison of NLP Systems, Online.10.18653/v1/2020.eval4nlp-1.9 ↗