Elamur Research⌕
☰
home / papers / 2024 / Analyzing Politeness in Arabic Tweets: A Preliminary Study

Analyzing Politeness in Arabic Tweets: A Preliminary Study

Summary

The authors address automatic politeness detection in Arabic social media, a task that has received little computational attention compared with English. They built a small corpus of 500 tweets taken from the ArSarcasm dataset, keeping only items tagged as MSA with positive sentiment. A linguistic expert manually labelled each tweet as polite or neutral, based on two criteria: lexical politeness markers and the pragmatic intent behind the message, such as appreciation, respect, love or admiration.

Two families of methods were compared. The first fine-tunes the Arabic encoders MarBERT and CamelBERT as baselines. The second uses zero-shot and few-shot prompting with three large language models: GPT-4o-mini, JAIS 30B Chat and Cohere Command. JAIS 30B Chat and GPT-4o-mini clearly outperformed Cohere Command, and adding more demonstrations did not improve results.

Error analysis showed that the models tend to over-predict politeness, miss the intent of a whole sentence, and are misled by emojis. As an early benchmark, the study highlights how much culture-aware resources matter for Arabic pragmatics.

Key points

  • Data and labels: 500 MSA tweets with positive sentiment from ArSarcasm, labelled for binary politeness (polite vs. neutral) using lexical markers plus pragmatic intention.
  • 7-shot results: JAIS 30B Chat ranked first, GPT-4o-mini reached 70.73% accuracy, and Cohere Command reached 59.15%.
  • 20-shot results: accuracy dropped slightly for all three models. JAIS 30B Chat scored 67.59%, GPT-4o-mini 66.39% and Cohere Command 58.87%.
  • Fine-tuned baselines: MarBERT labelled every tweet as polite, a degenerate output. CamelBERT matched the manual labels except for about 12% of cases.
  • Stated limitations: the corpus is small and MSA-only, a single annotator introduces subjectivity, and binary labels may oversimplify politeness. Planned future work includes larger datasets and the integration of cultural and pragmatic knowledge.

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: Linguistics & languages

Cited by

4 papers of our library cite it (of the 140673 whose bibliographies we read); 1 of them have a page here, 3 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{alkhalifa2024analyzing,
  title = {{Analyzing Politeness in Arabic Tweets: A Preliminary Study}},
  author = {Al-Khalifa, Hend and Ghezaiel, Nadia and Bounnit, Maria},
  booktitle = {Proceedings of the 7th International Conference on Natural Language and Speech Processing (ICNLSP 2024)},
  year = {2024},
  url = {https://aclanthology.org/2024.icnlsp-1.36/}
}

References

19 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Alaearji, A. A., & Buraihi, R. H. (2021). The Principle of Kindness and Politeness in Arabic Linguistic Thought. International Journal on Humanities and Social Sciences, 27, 42–50. https://doi.org/10.33193/IJoHSS.27.2021.336 ↗
  2. Ameri, M., Zeighami, A., & Mirahmadi, S. R. (2023). A study of the polite method in the Arabic language according to Brown and Levinson’s theory of politeness. Studies on Arabic Language and Literature, 13 (36), 61–88. https://doi.org/10.22075/lasem.2022.23600.1286 ↗
  3. Aubakirova, M., & Bansal, M. (2016). Interpreting Neural Networks to Improve Politeness Comprehension. In J. Su, K. Duh, & X. Carreras (Eds.), Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing (pp. 2035– 2041). Association for Computational Linguistics. https://doi.org/10.18653/v1/D16-1216 ↗ our card
  4. Brown, P., & Levinson, S. C. (1987). Politeness: Some Universals in Language Usage. Cambridge University Press. find our card
  5. Danescu-Niculescu-Mizil, C., Sudhof, M., Jurafsky, D., Leskovec, J., & Potts, C. (2013). A computational approach to politeness with application to social factors. the 51st Annual Meeting of the Association for Computational Linguistics. find our card
  6. Dasgupta, T., Sinha, M., & Geetha Praveen, C. (2023). Graph Induced Transformer Network for Detection of Politeness and Formality in Text. Companion Proceedings of the ACM Web Conference 2023, 221– 224. https://doi.org/10.1145/3543873.3587352 ↗
  7. Feely, W., Hasler, E., & de Gispert, A. (2019). Controlling Japanese Honorifics in English-to-Japanese Neural Machine Translation. In T. Nakazawa, C. Ding, R. Dabre, A. Kunchukuttan, N. Doi, Y. Oda, O. Bojar, S. Parida, I. Goto, & H. Mino (Eds.), Proceedings of the 6th Workshop on Asian Translation (pp. 45–53). Association for Computational Linguistics. https://doi.org/10.18653/v1/D19-5203 ↗
  8. Firdaus, M., Ekbal, A., & Bhattacharyya, P. (2020). Incorporating Politeness across Languages in Customer Care Responses: Towards building a Multilingual Empathetic Dialogue Agent. In N. Calzolari, F. Béchet, P. Blache, K. Choukri, C. Cieri, T. Declerck, S. Goggi, H. Isahara, B. Maegaard, J. Mariani, H. Mazo, A. Moreno, J. Odijk, & S. Piperidis (Eds.), Proceedings of the Twelfth Language Resources and Evaluation Conference (pp. 4172–4182). European Language Resources Association. https://aclanthology.org/2020.lrec-1.514 ↗
  9. Fu, L., Fussell, S., & Danescu-Niculescu-Mizil, C. (2020). Facilitating the Communication of Politeness through Fine-Grained Paraphrasing. In B. Webber, T. Cohn, Y. He, & Y. Liu (Eds.), Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) (pp. 5127–5140). Association for Computational Linguistics. https://doi.org/10.18653/v1/2020.emnlp-main.416 ↗ our card
  10. Golchha, H., Firdaus, M., Ekbal, A., & Bhattacharyya, P. (2019). Courteously Yours: Inducing courteous behavior in Customer Care responses using Reinforced Pointer Generator Network. In J. Burstein, C. Doran, & T. Solorio (Eds.), Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) (pp. 851–860). Association for Computational Linguistics. https://doi.org/10.18653/v1/N19-1091 ↗
  11. Hoffman, E. R., McDonald, D. W., & Zachry, M. (2017). Evaluating a Computational Approach to Labeling Politeness: Challenges for the Application of Machine Classification to Social Computing Data. Proc. ACM Hum.-Comput. Interact., 1(CSCW), 52:1-52:14. https://doi.org/10.1145/3134687 ↗
  12. Kumar, R. (2021). Towards automatic identification of linguistic politeness in Hindi texts (arXiv:2111.15268). arXiv. https://doi.org/10.48550/arXiv.2111.15268 ↗
  13. Li, C., Pang, B., Wang, W., Hu, L., Gordon, M., Marinova, D., Balducci, B., & Shang, Y. (2023). How Well Can Language Models Understand Politeness? 2023 IEEE Conference on Artificial Intelligence (CAI), 230–231. https://doi.org/10.1109/CAI54212.2023.00106 ↗
  14. Mishra, K., Firdaus, M., & Ekbal, A. (2022). Please be polite: Towards building a politeness adaptive dialogue system for goal-oriented conversations. Neurocomputing, 494, 242–254. https://doi.org/10.1016/j.neucom.2022.04.029 ↗
  15. Mishra, K., Firdaus, M., & Ekbal, A. (2023). Predicting Politeness Variations in Goal-Oriented Conversations. IEEE Transactions on Computational Social Systems, 10(3), 1095–1104. IEEE Transactions on Computational Social Systems. https://doi.org/10.1109/TCSS.2022.3156580 ↗
  16. Niu, T., & Bansal, M. (2018). Polite Dialogue Generation Without Parallel Data. Transactions of the Association for Computational Linguistics, 6, 373– 389. https://doi.org/10.1162/tacl_a_00027 ↗
  17. Sennrich, R., Haddow, B., & Birch, A. (2016). Controlling Politeness in Neural Machine Translation via Side Constraints. In K. Knight, A. Nenkova, & O. Rambow (Eds.), Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (pp. 35–40). Association for Computational Linguistics. https://doi.org/10.18653/v1/N16-1005 ↗ our card
  18. Wang, Y.-C., Papangelis, A., Wang, R., Feizollahi, Z., Tur, G., & Kraut, R. (2020). Can You be More Social? Injecting Politeness and Positivity into Task-Oriented Conversational Agents (arXiv:2012.14653). arXiv. https://doi.org/10.48550/arXiv.2012.14653 ↗
  19. Ziems, C., Held, W., Shaikh, O., Chen, J., Zhang, Z., & Yang, D. (2024). Can Large Language Models Transform Computational Social Science? Computational Linguistics, 50(1), 237–291. https://doi.org/10.1162/coli_a_00502 ↗ our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/khalifa-2024-analyzing-politeness-arabic-tweets-preliminary/