Emotional Tone Detection in Hate Speech Using Machine Learning and NLP: Methods, Challenges, and Future Directions—A Systematic Review
Source abstract
Abstract
Hate speech is a form of communicative expression that promotes or incites unjustified violence. The increase in hate speech on social media has prompted the development of automated tools for its detection, especially those that integrate emotional tone analysis. This study presents a systematic review of the literature, employing a combination of PRISMA and PICOS methodologies to identify the most used Machine Learning techniques and Natural Language Processing emotion classification in hostile messages. It also seeks to determine which models and tools predominate in the analyzed studies. The findings highlight LLaMA 2 and HingRoBERTa, achieving F1 scores of 100% and 98.45%, respectively. Furthermore, key challenges are identified, including linguistic bias, language ambiguity, and the high computational demands of some models. This review contributes an updated overview of the state of the art, highlighting the need for more inclusive, efficient, and interpretable approaches to improve automated moderation on digital platforms. Additionally, it includes techniques, methods, and future directions in this topic.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Hate Speech and Cyberbullying Detection · Emotion and Mood Recognition · Sentiment Analysis and Opinion Mining
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Computer Science, Universidad de las Fuerzas Armadas (ESPE), Av. General Rumiñahui, Quito 171103, Ecuador
Funding
Universidad de las Fuerzas Armadas ESPE
References (45)
Show source-provided references
1. A comprehensive review on automatic hate speech detection in the age of the transformer10.1007/s13278-024-01361-3 ↗
2. Zhang, F., Chen, J., Tang, Q., and Tian, Y. (2024). Evaluation of emotion classification schemes in social media text: An annotation-based approach. BMC Psychol., 12.10.1186/s40359-024-02008-w ↗
3. Finding hate speech with auxiliary emotion detection from self-training multi-label learning perspective10.1016/j.inffus.2023.03.015 ↗
4. Multi-modal Twitter Data Analysis for Identifying Offensive Posts Using a Deep Cross-Attention–based Transformer Framework10.1145/3713077 ↗
5. Baruah, A., Wahlang, L., Jyrwa, F., Shadap, F., Barbhuiya, F., and Dey, K. (2024). Abusive Language Detection in Khasi Social Media Comments. ACM Trans. Asian Low-Resour. Lang. Inf. Process.10.1145/3664285 ↗
6. Unlocking language barriers: Assessing pre-trained large language models across multilingual tasks and unveiling the black box with Explainable Artificial Intelligence10.1016/j.engappai.2025.110136 ↗
7. An explanation framework and method for AI-based text emotion analysis and visualisation10.1016/j.dss.2023.114121 ↗
8. Declaración PRISMA 2020: Una guía actualizada para la publicación de revisiones sistemáticas10.1016/j.recesp.2021.06.016 ↗
9. Using the full PICO model as a search tool for systematic reviews resulted in lower recall for some PICO elements10.1016/j.jclinepi.2020.07.005 ↗
10. Ransomware Detection with Machine Learning: Techniques, Challenges, and Future Directions—A Systematic Review
11. Adversarial examples: A survey of attacks and defenses in deep learning-enabled cybersecurity systems10.1016/j.eswa.2023.122223 ↗
12. Benavides-Astudillo, E., Fuertes, W., Sanchez-Gordon, S., Nuñez-Agurto, D., and Rodríguez-Galán, G. (2023). A Phishing-Attack-Detection Model Using Natural Language Processing and Deep Learning. Appl. Sci., 13.10.3390/app13095275 ↗
13. Forming research questions10.1016/j.jclinepi.2006.06.006 ↗
14. Leveraging distant supervision and deep learning for twitter sentiment and emotion classification10.1007/s10844-024-00845-0 ↗
15. Cyberbullying Detection Based on Emotion10.1109/ACCESS.2023.3280556 ↗
16. FADOHS: Framework for Detection and Integration of Unstructured Data of Hate Speech on Facebook Using Sentiment and Emotion Analysis10.1109/ACCESS.2022.3151098 ↗
17. A multi-task learning framework for politeness and emotion detection in dialogues for mental health counselling and legal aid10.1016/j.eswa.2023.120025 ↗
18. Language augmentation approach for code-mixed text classification10.1016/j.nlp.2023.100042 ↗
19. G-BERT: An Efficient Method for Identifying Hate Speech in Bengali Texts on Social Media10.1109/ACCESS.2023.3299021 ↗
20. Koufakou, A., Garciga, J., Paul, A., Morelli, J., and Frank, C. (November, January 31). Automatically Classifying Emotions Based on Text: A Comparative Exploration of Different Datasets. Proceedings of the 2022 IEEE 34th International Conference on Tools with Artificial Intelligence (ICTAI), Macao, China.10.1109/ICTAI56018.2022.00056 ↗
21. An efficient CNN-LSTM model for sentiment detection in #BlackLivesMatter10.1016/j.eswa.2021.116256 ↗
22. Mittal, U. (2023, January 26–27). Detecting Hate Speech Utilizing Deep Convolutional Network and Transformer Models. Proceedings of the 2023 International Conference on Electrical, Electronics, Communication and Computers (ELEXCOM), Roorkee, India.10.1109/ELEXCOM58812.2023.10370502 ↗
23. Leveraging bilingual-view parallel translation for code-switched emotion detection with adversarial dual-channel encoder10.1016/j.knosys.2021.107436 ↗
24. TurkishBERTweet: Fast and reliable large language model for social media analysis10.1016/j.eswa.2024.124737 ↗
25. A review of multimodal-based emotion recognition techniques for cyberbullying detection in online social media platforms10.1007/s00521-024-10371-3 ↗
26. Alvarez-Gonzalez, N., Kaltenbrunner, A., and Gómez, V. (2021, January 16–20). Uncovering the Limits of Text-based Emotion Detection. Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2021, Virtual.10.18653/v1/2021.findings-emnlp.219 ↗
27. Bashynska, I., Sarafanov, M., and Manikaeva, O. (2024). Research and Development of a Modern Deep Learning Model for Emotional Analysis Management of Text Data. Appl. Sci., 14.10.3390/app14051952 ↗
28. Chakraborty, P., Nawar, F., and Chowdhury, H.A. (2021). Sentiment Analysis of Bengali Facebook Data Using Classical and Deep Learning Approaches. Innovation in Electrical Power Engineering, Communication, and Computing Technology, Springer.10.1007/978-981-16-7076-3_19 ↗
29. Improving Text Emotion Detection Through Comprehensive Dataset Quality Analysis10.1109/ACCESS.2024.3491856 ↗
30. Enhancing sentiment analysis with distributional emotion embeddings10.1016/j.neucom.2025.129822 ↗
31. Spanish MTLHateCorpus 2023: Multi-task learning for hate speech detection to identify speech type, target, target group and intensity10.1016/j.csi.2025.103990 ↗
32. Emotion Detection in Hinglish(Hindi+English) Code-Mixed Social Media Text10.1016/j.procs.2020.04.144 ↗
33. Sohail, T., Aiman, A., Hashmi, E., Imran, A.S., Daudpota, S.M., and Yayilgan, S.Y. (2024, January 9–10). Hate Speech Detection in Code-Mixed Datasets Using Pretrained Embeddings and Transformers. Proceedings of the 2024 International Conference on Frontiers of Information Technology (FIT), Islamabad, Pakistan.10.1109/FIT63703.2024.10838452 ↗
34. Combining profile features for offensiveness detection on Spanish social media10.1016/j.eswa.2025.126705 ↗
35. Alaeddini, M. (2024). Emotion Detection in Reddit: Comparative Study of Machine Learning and Deep Learning Techniques. arXiv.
36. Integrating personalized and contextual information in fine-grained emotion recognition in text: A multi-source fusion approach with explainability10.1016/j.inffus.2025.102966 ↗
37. Enhancement of a multi-dialectal sentiment analysis system by the detection of the implied sarcastic features10.1016/j.knosys.2021.107232 ↗
38. Lecourt, F., Croitoru, M., and Todorov, K. (May, January 28). ‘Only ChatGPT gets me’: An Empirical Analysis of GPT versus other Large Language Models for Emotion Detection in Text. Proceedings of the WWW ’25: Companion Proceedings of the ACM on Web Conference 2025, Sydney, NSW, Australia.10.1145/3701716.3718375 ↗
39. Revisiting Sentiment Analysis for Software Engineering in the Era of Large Language Models10.1145/3697009 ↗
40. Youth language and emerging slurs: Tackling bias in BERT-based hate speech detection10.1007/s43681-025-00701-z ↗
41. Detecting misogyny in Spanish tweets. An approach based on linguistics features and word embeddings10.1016/j.future.2020.08.032 ↗
42. Fuzzy rough nearest neighbour methods for detecting emotions, hate speech and irony10.1016/j.ins.2023.01.054 ↗
43. Andrade, R.O., Fuertes, W., Cazares, M., Ortiz-Garcés, I., and Navas, G. (2022). An Exploratory Study of Cognitive Sciences Applied to Cybersecurity. Electronics, 11.10.3390/electronics11111692 ↗
44. Current State of Text Sentiment Analysis from Opinion to Emotion Mining10.1145/3057270 ↗
45. Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow, C.D., Shamseer, L., Tetzlaff, J.M., Akl, E.A., and Brennan, S.E. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372.10.1136/bmj.n71 ↗