Enhanced Hate Speech Detection in Indonesian-English Code-Mixed Texts Using XLM-RoBERTa
Abstract
The prevalence of hate speech on digital platforms presents significant challenges, particularly in multilingual communities where code-mixing complicates detection. This study explores the use of XLMRoBERTa, a transformer-based model with robust multilingual capabilities, to detect hate speech within code-mixed texts, focusing on Indonesian-English code-mixing. Traditional hate speech detection models rely on single-language datasets, limiting their effectiveness in such environments. We employ a dataset consisting of Indonesian, English, and code-mixed Indonesian-English texts to evaluate XLMRoBERTa's performance. The dataset comprises 24,844 training samples, 2,760 test samples, and an additional 100 supplementary samples. Key hyperparameters included a batch size of 16 and 32, with a learning rate ranging from 1e-5 to 5e-5. The model achieved near-perfect accuracy (99.6%) on the primary test set and demonstrated strong generalization across realistic supplementary data, achieving an F1-score of 90.94%. These findings underscore the model's potential for application in complex linguistic contexts, contributing to the development of effective code-mixed hate speech detection.References
J. B. Walther, “Social media and online hate,” Curr. Opin. Psychol., vol. 45, no. January, 2022, doi: 10.1016/j.copsyc.2021.12.010.
K. Sreelakshmi, B. Premjith, and K. P. Soman, “Detection of Hate Speech Text in Hindi-English Code-mixed Data,” Procedia Comput. Sci., vol. 171, no. 2019, pp. 737–744, 2020, doi: 10.1016/j.procs.2020.04.080.
E. W. Pamungkas, A. Fatmawati, Y. S. Nugroho, D. Gunawan, and E. Sudarmilah, “Hate Speech Detection in Code-Mixed Indonesian Social Media: Exploiting Multilingual Languages Resources,” in 2022 Seventh International Conference on Informatics and Computing (ICIC), IEEE, Dec. 2022, pp. 1–5. doi: 10.1109/ICIC56845.2022.10006940.
M. S. Jahan and M. Oussalah, “A systematic review of hate speech automatic detection using natural language processing,” Neurocomputing, vol. 546, p. 126232, Aug. 2023, doi: 10.1016/j.neucom.2023.126232.
L. Xu, J. Zeng, and S. Chen, “yasuo at HASOC2020: Fine-tune XML-RoBERTa for Hate Speech Identification,” 2020.
X. Ou and H. Li, “YNU@Dravidian-CodeMix-FIRE2020: XLM-RoBERTa for Multi-language Sentiment Analysis,” 2020.
T. Tita, Q. Mary, and A. Zubiaga, “Cross-lingual Hate Speech Detection using Transformer Models”, doi: 10.48550/arXiv.2111.00981.
T. Leburu-Dingalo, K. Johannes Ntwaagae, N. Peace Motlogelwa, E. Thuma, and M. Mudongo, “Application of XLM-RoBERTa for Multi-Class Classification of Conversational Hate Speech,” 2022.
S. Wang, J. Liu, X. Ouyang, and Y. Sun, “Galileo at SemEval-2020 Task 12: Multi-lingual Learning for Offensive Language Identification using Pre-trained Language Models,” 14th Int. Work. Semant. Eval. SemEval 2020 - co-located 28th Int. Conf. Comput. Linguist. COLING 2020, Proc., pp. 1448–1455, 2020, doi: 10.18653/v1/2020.semeval-1.189.
A. Conneau et al., “Unsupervised Cross-lingual Representation Learning at Scale,” Nov. 2019.
Y. Liu et al., “RoBERTa: A robustly optimized BERT pretraining approach,” arXiv, no. 1, 2019.
J. Devlin, M. W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding,” in NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference, 2019.
Anonymous, “XLM-RoBERTa GitHub,” https://github.com/huggingface/transformers/blob/main/docs/source/en/model_doc/xlm-roberta.md.
Y. Meng et al., “Representation Deficiency in Masked Language Modeling,” Feb. 2023.
Anonymous, “XLM-RoBERTa HuggingFace,” https://huggingface.co/docs/transformers/v4.46.0/en/model_doc/xlm-roberta#transformers.TFXLMRobertaForSequenceClassification.
X. Amatriain, A. Sankar, J. Bing, P. K. Bodigutla, T. J. Hazen, and M. Kazi, “Transformer models: an introduction and catalog,” Feb. 2023.
G. Lample and A. Conneau, “Cross-lingual Language Model Pretraining,” Jan. 2019.
DOI:
https://doi.org/10.31449/inf.v49i21.7713Downloads
Published
How to Cite
Issue
Section
License
Authors retain copyright in their work. By submitting to and publishing with Informatica, authors grant the publisher (Slovene Society Informatika) the non-exclusive right to publish, reproduce, and distribute the article and to identify itself as the original publisher.
All articles are published under the Creative Commons Attribution license CC BY 3.0. Under this license, others may share and adapt the work for any purpose, provided appropriate credit is given and changes (if any) are indicated.
Authors may deposit and share the submitted version, accepted manuscript, and published version, provided the original publication in Informatica is properly cited.
