Elamur Research⌕
☰
home / papers / 2022 / Hierarchical Aggregation of Dialectal Data for Arabic Dialect Identification

Hierarchical Aggregation of Dialectal Data for Arabic Dialect Identification

Summary

Arabic dialect identification (DID) resources have grown quickly, but each corpus labels its texts differently. Some use regions, others countries, provinces or cities, and the label sets do not line up. This makes it hard to combine or compare data. The authors propose a single three-level hierarchy (region → country → city) for labeling dialectal Arabic. They write dataset-specific code that maps the original labels of 29 corpora into this hierarchy. Using the merged data, they train character- and word-based n-gram language models at each of the three levels. To test whether aggregation pays off, they feed these LMs into an established, strong DID approach and evaluate on a standard city-level test set, reporting that the aggregated models are useful. The mapping code and the LMs are released publicly. The work matters because it turns many scattered dialect corpora into one reusable resource. It also offers a template for other dialect families where every corpus has its own labeling conventions.

Key points

  • The unified schema has three levels: region, country and city. Source labels at other granularities, including provinces, are mapped into it.
  • 29 corpora are harmonized, among them MADAR (25 cities), NADI (21 countries, 100 provinces), AOC, Curras, Gumar, Habibi, PADIC, QADI, Shami, SUAR, YouDACC and several LDC transcript sets.
  • Aggregated n-gram LMs are built at region, country and city levels, in both character and word spaces.
  • The LMs are evaluated by extending a well-established state-of-the-art Arabic DID system on a standard city-level test set.
  • Arabizi (romanized Arabic) is explicitly not modeled. The code and LMs are available at github.com/CAMeL-Lab/HierarchicalArabicDialectID.

Interesting ideas

  • I1. Each source corpus gets its own small adapter that rewrites its labels into one shared hierarchy, so heterogeneous datasets can be pooled without relabeling by hand. — For Mova (tree): UD dialect conversion could use per-treebank converters into one canonical label hierarchy, keeping each mapping explicit and auditable.
  • I2. A hierarchical label set lets coarsely labeled data still count: a text tagged only by region can train region-level models even when it says nothing about the city. — For Mova (learn): silver data where witness parsers agree only on a coarse category, such as the UPOS class but not the feature bundle, could still feed training at the level where agreement holds.
  • I3. One n-gram LM per class, over both characters and words, gives transparent per-class likelihood scores that capture spelling variation as well as vocabulary. — For Mova (tokens): per-variety character LMs could flag British vs. American spelling or informal register before tagging, with scores that explain the decision.
  • I4. The value of a new resource is shown by plugging it into an existing strong system rather than building a new model, which isolates the contribution of the data itself. — For Mova (learn): new knowledge sources or silver sets could be evaluated by adding them to the current parser or reader and measuring the change, with the unchanged system as the control.
  • I5. Because short vowels are often unwritten and speakers mix dialect with MSA, many words look identical across varieties, so the paper stresses using context rather than isolated words. — For Mova (frames): ambiguous surface forms should be resolved by looking at their surrounding structure rather than by single-word lookups.

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

17 papers of our library cite it (of the 140884 whose bibliographies we read); 2 of them have a page here, 15 not yet.

Cite

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

@inproceedings{baimukan2022hierarchical,
  title = {{Hierarchical Aggregation of Dialectal Data for Arabic Dialect Identification}},
  author = {Baimukan, Nurpeiis and Bouamor, Houda and Habash, Nizar},
  booktitle = {Proceedings of the Thirteenth Language Resources and Evaluation Conference},
  year = {2022},
  doi = {10.63317/2wdgcvd7eoho},
  url = {https://doi.org/10.63317/2wdgcvd7eoho}
}

References

69 references, 20 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Abdul-Mageed, M., Zhang, C., Bouamor, H., and Habash, N. (2020). NADI 2020: The First Nuanced Arabic Dialect Identification Shared Task. In Proceedings of the Fifth Arabic Natural Language Processing Workshop (WANLP 2020), Barcelona, Spain. find
  2. Abdul-Mageed, M., Zhang, C., Elmadany, A., Bouamor, H., and Habash, N. (2021). NADI 2021: The second nuanced Arabic dialect identification shared task. In Proceedings of the Sixth Arabic Natural Language Processing Workshop, pages 244– 259, Kyiv, Ukraine (Virtual), April. Association for Computational Linguistics. find
  3. Al-Twairesh, N., Al-Matham, R. N., Madi, N., Almugren, N., Al-Aljmi, A.-H., Alshalan, S., Alshalan, R., Alrumayyan, N., Al-Manea, S., Bawazeer, S., Al-Mutlaq, N., Almanea, N., Huwaymil, W. B., Alqusair, D., Alotaibi, R., Al-Senaydi, S., and Alfutamani, A. (2018). Suar: Towards building a corpus for the saudi dialect. In ACLING. find
  4. Ali, A., Chowdhury, S., Hussein, A., and Hifny, Y. (2021). Arabic code-switching speech recognition using monolingual data. arXiv preprint arXiv:2107.01573. ↗ our card
  5. Althobaiti, M. J. (2020). Automatic Arabic dialect identification systems for written texts: A survey. arXiv preprint arXiv:2009.12622. ↗ our card
  6. Bies, A., Song, Z., Maamouri, M., Grimes, S., Lee, H., Wright, J., Strassel, S., Habash, N., Eskander, R., and Rambow, O. (2014). Transliteration of Arabizi into Arabic Orthography: Developing a Parallel Annotated Arabizi-Arabic Script SMS/Chat Corpus. In Proceedings of the Workshop for Arabic Natural Language Processing (WANLP), Doha, Qatar. find
  7. Bouamor, H., Habash, N., Salameh, M., Zaghouani, W., Rambow, O., Abdulrahim, D., Obeid, O., Khalifa, S., Eryani, F., Erdmann, A., and Oflazer, K. (2018). The MADAR Arabic Dialect Corpus and Lexicon. In Proceedings of the Language Resources and Evaluation Conference (LREC), Miyazaki, Japan. find
  8. Bouamor, H., Hassan, S., and Habash, N. (2019). The MADAR shared task on Arabic fine-grained dialect identification. In Proceedings of the Fourth Arabic Natural Language Processing Workshop, pages 199–207, Florence, Italy, August. Association for Computational Linguistics. find our card
  9. Darwish, K. (2014). Arabizi Detection and Conversion to Arabic. In Proceedings of the Workshop for Arabic Natural Language Processing (WANLP), pages 217–224, Doha, Qatar. find
  10. El-Haj, M., Rayson, P., and Aboelezz, M. (2018). Arabic Dialect Identification in the Context of Bivalency and Code-Switching. In Proceedings of the Language Resources and Evaluation Conference (LREC), Miyazaki, Japan. find
  11. Elfardy, H., Al-Badrashiny, M., and Diab, M. (2013). Code Switch Point Detection in Arabic. In Proceedings of the Conference on Application of Natural Language to Information Systems (NLDB), MediaC-ity, UK. find
  12. Etman, A. and Beex, L. (2015). Language and Dialect Identification: A Survey. In Proceedings of the Intelligent Systems Conference (IntelliSys), London, UK. find
  13. Habash, N., Soudi, A., and Buckwalter, T. (2007). On Arabic Transliteration. In A. van den Bosch et al., editors, Arabic Computational Morphology: Knowledge-based and Empirical Methods, pages 15–22. Springer, Netherlands. find
  14. Habash, N., Rambow, O., Diab, M., and Kanjawi-Faraj, R. (2008). Guidelines for Annotation of Arabic Dialectness. In Proceedings of the Workshop on HLT & NLP within the Arabic World, Marrakech, Morocco. find
  15. Habash, N., Eryani, F., Khalifa, S., Rambow, O., Abdulrahim, D., Erdmann, A., Faraj, R., Zaghouani, W., Bouamor, H., Zalmout, N., Hassan, S., shargi, F. A., Alkhereyf, S., Abdulkareem, B., Eskander, R., Salameh, M., and Saddiki, H. (2018). Unified guidelines and resources for Arabic dialect orthography. In Proceedings of the Language Resources and Evaluation Conference (LREC), Miyazaki, Japan. find our card
  16. Habash, N. Y. (2010). Introduction to Arabic natural language processing. Morgan & Claypool Publishers. find
  17. Heafield, K. (2011). KenLM: Faster and Smaller Language Model Queries. In Proceedings of the Workshop on Statistical Machine Translation (WMT), pages 187–197. find our card
  18. Inoue, G., Alhafni, B., Baimukan, N., Bouamor, H., and Habash, N. (2021). The interplay of variant, size, and task type in Arabic pre-trained language models. In Proceedings of the Sixth Arabic Natural Language Processing Workshop, pages 92–104, Kyiv, Ukraine (Virtual), April. Association for Computational Linguistics. find our card
  19. Jarrar, M., Habash, N., Alrimawi, F., Akra, D., and Zalmout, N. (2016). Curras: an annotated corpus for the Palestinian Arabic dialect. Language Resources and Evaluation, pages 1–31. find
  20. Jauhiainen, T., Lui, M., Zampieri, M., Baldwin, T., and Lindén, K. (2019). Automatic language identification in texts: A survey. Journal of Artificial Intelligence Research, 65, Aug. find our card
  21. Malmasi, S., Zampieri, M., Ljubešić, N., Nakov, P., Ali, A., and Tiedemann, J. (2016). Discriminating between Similar Languages and Arabic Dialect Identification: A Report on the Third DSL Shared Task. In Proceedings of the Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial), pages 1–14, Osaka, Japan. find our card
  22. McNemar, Q. (1947). Note on the sampling error of the difference between correlated proportions or percentages. Psychometrika, 12(2):153–157. find
  23. Obeid, O., Zalmout, N., Khalifa, S., Taji, D., Oudah, M., Alhafni, B., Inoue, G., Eryani, F., Erdmann, A., and Habash, N. (2020). CAMeL tools: An open source python toolkit for Arabic natural language processing. In Proceedings of the 12th Language Resources and Evaluation Conference, pages 7022– 7032, Marseille, France, May. European Language Resources Association. find our card
  24. Salameh, M., Bouamor, H., and Habash, N. (2018). Fine-grained Arabic dialect identification. In Proceedings of the International Conference on Computational Linguistics (COLING), pages 1332–1344, Santa Fe, New Mexico, USA. find our card
  25. Shoufan, A. and Alameri, S. (2015). Natural language processing for dialectical Arabic: A survey. In Proceedings of the Second Workshop on Arabic Natural Language Processing, pages 36–48, Beijing, China, July. Association for Computational Linguistics. find
  26. Smaı̈li, K., Abbas, M., Meftouh, K., and Harrat, S. (2014). Building resources for Algerian Arabic dialects. In Proceedings of the Conference of the International Speech Communication Association (Interspeech). find
  27. Takezawa, T., Kikui, G., Mizushima, M., and Sumita, E. (2007). Multilingual Spoken Language Corpus Development for Communication Research. Computational Linguistics and Chinese Language Processing, 12(3):303–324. find
  28. Zaidan, O. F. and Callison-Burch, C. (2011). The Arabic Online Commentary Dataset: an Annotated Dataset of Informal Arabic With High Dialectal Content. In Proceedings of the Conference of the Association for Computational Linguistics (ACL), pages 37–41. find our card
  29. Zaidan, O. F. and Callison-Burch, C. (2012). Arabic dialect identification. Computational Linguistics, 40(1):171–202. find our card
  30. Zampieri, M., Malmasi, S., Ljubešić, N., Nakov, P., Ali, A., Tiedemann, J., Scherrer, Y., and Aepli, N. (2017). Findings of the VarDial Evaluation Campaign 2017. In Proceedings of the Workshop on NLP for Similar Languages, Varieties and Dialects (Var-Dial), pages 1–15, Valencia, Spain. find our card
  31. Zampieri, M., Malmasi, S., Nakov, P., Ali, A., Shon, S., Glass, J., Scherrer, Y., Samardžić, T., Ljubešić, N., Tiedemann, J., van der Lee, C., Grondelaers, S., Oostdijk, N., Speelman, D., van den Bosch, A., Kumar, R., Lahiri, B., and Jain, M. (2018). Language identification and morphosyntactic tagging: The second VarDial evaluation campaign. In Proceedings of the Fifth Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial 2018), pages 1–17, Santa Fe, New Mexico, USA, August. Association for Computational Linguistics. 8. find our card
  32. Language Resource References Abdelali, A., Mubarak, H., Samih, Y., Hassan, S., and Darwish, K. (2020). Arabic dialect identification in the wild. find our card
  33. Abdul-Mageed, M., Alhuzali, H., and Elaraby, M. (2018). You tweet what you speak: A city-level dataset of Arabic dialects. In Proceedings of the Language Resources and Evaluation Conference (LREC), Miyazaki, Japan. find
  34. Abdul-Mageed, M., Zhang, C., Bouamor, H., and Habash, N. (2020). NADI 2020: The First Nuanced Arabic Dialect Identification Shared Task. In Proceedings of the Fifth Arabic Natural Language Processing Workshop (WANLP 2020), Barcelona, Spain. find
  35. Abu Kwaik, K., Saad, M., Chatzikyriakidis, S., and Dobnik, S. (2018). Shami: A corpus of Levantine Arabic dialects. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018), Miyazaki, Japan, May. European Language Resources Association (ELRA). find
  36. Al-Sabbagh, R. and Girju, R. (2012). YADAC: Yet another Dialectal Arabic Corpus. In Proceedings of the Language Resources and Evaluation Conference (LREC), pages 2882–2889, Istanbul, Turkey. find
  37. Al-Twairesh, N., Al-Matham, R. N., Madi, N., Almugren, N., Al-Aljmi, A.-H., Alshalan, S., Alshalan, R., Alrumayyan, N., Al-Manea, S., Bawazeer, S., Al-Mutlaq, N., Almanea, N., Huwaymil, W. B., Alqusair, D., Alotaibi, R., Al-Senaydi, S., and Alfutamani, A. (2018). Suar: Towards building a corpus for the saudi dialect. In ACLING. find
  38. Almeman, K. and Lee, M. (2013). Automatic building of Arabic multi dialect text corpora by bootstrapping dialect words. In Proceedings of the International Conference on Communications, Signal Processing, and their Applications (ICCSPA), pages 1–6. find
  39. Appen Pty Ltd. (2006a). Gulf Arabic Conversational Telephone Speech, Transcripts LDC2006T15. find
  40. Appen Pty Ltd. (2006b). Iraqi Arabic Conversational Telephone Speech, Transcripts LDC2006T16. find
  41. Appen Pty Ltd. (2007). Levantine Arabic Conversational Telephone Speech, Transcripts LDC2007T01. find
  42. BBN Technologies, R., Linguistic Data Consortium, and Sakhr Software. (2012). Arabic-Dialect/English Parallel Text LDC2012T09. find
  43. BBN Technologies. (2005). BBN/AUB DARPA Babylon Levantine Arabic Speech and Transcripts LDC2005S08. find
  44. Bouamor, H., Habash, N., and Oflazer, K. (2014). A multidialectal parallel corpus of Arabic. In Proceedings of the Language Resources and Evaluation Conference (LREC), Reykjavik, Iceland. find our card
  45. Bouamor, H., Habash, N., Salameh, M., Zaghouani, W., Rambow, O., Abdulrahim, D., Obeid, O., Khalifa, S., Eryani, F., Erdmann, A., and Oflazer, K. (2018). The MADAR Arabic Dialect Corpus and Lexicon. In Proceedings of the Language Resources and Evaluation Conference (LREC), Miyazaki, Japan. find
  46. Bouamor, H., Hassan, S., and Habash, N. (2019). The MADAR shared task on Arabic fine-grained dialect identification. In Proceedings of the Fourth Arabic Natural Language Processing Workshop, pages 199–207, Florence, Italy, August. Association for Computational Linguistics. find our card
  47. Chen, S., Fore, D., Strassel, S., Lee, H., and Wright, J. (2017). Bolt egyptian arabic sms/chat and transliteration ldc2017t07. find
  48. Cotterell, R. and Callison-Burch, C. (2014). A multi-dialect, multi-genre corpus of informal written Arabic. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC’14), pages 241–245, Reykjavik, Iceland, May. European Language Resources Association (ELRA). find
  49. Diab, M., Habash, N., Rambow, O., AlTantawy, M., and Benajiba, Y. (2010). COLABA: Arabic Dialect Annotation and Processing. In Proceedings of the Workshop on Language Resources and Human Language Technology for Semitic Languages. find
  50. El-Haj, M. (2020). Habibi - a multi dialect multi national Arabic song lyrics corpus. In Proceedings of the 12th Language Resources and Evaluation Conference, pages 1318–1326, Marseille, France, May. European Language Resources Association. find our card
  51. Elfardy, H., Al-Badrashiny, M., and Diab, M. (2014). Aida: Identifying code switching in informal Arabic text. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 94–101, Doha, Qatar. find
  52. Gadalla, H., Kilany, H., Arram, H., Yacoub, A., El-Habashi, A., Shalaby, A., Karins, K., Rowson, E., MacIntyre, R., Kingsbury, P., Graff, D., and McLemore, C. (1997). CALLHOME Egyptian Arabic Transcripts LDC97T19. find
  53. Jarrar, M., Habash, N., Alrimawi, F., Akra, D., and Zalmout, N. (2016). Curras: an annotated corpus for the Palestinian Arabic dialect. Language Resources and Evaluation, pages 1–31. find
  54. Khalifa, S., Habash, N., Abdulrahim, D., and Hassan, S. (2016). A Large Scale Corpus of Gulf Arabic. In Proceedings of the Language Resources and Evaluation Conference (LREC), Portorož, Slovenia. find
  55. Khalifa, S., Habash, N., Eryani, F., Obeid, O., Abdulrahim, D., and Al Kaabi, M. (2018). A morphologically annotated corpus of Emirati Arabic. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018), Miyazaki, Japan, May. European Language Resources Association (ELRA). find
  56. Linguistic Data Consortium. (2002). CALL-HOME Egyptian Arabic Transcripts Supplement LDC2002T38. find
  57. Maamouri, M., Buckwalter, T., and Jin, H. (2005). Levantine Arabic QT Training Data Set 4 (Speech + Transcripts) LDC2005S14. find
  58. Maamouri, M., Buckwalter, T., Graff, D., and Jin, H. (2006). Levantine Arabic QT Training Data Set 5, Transcripts LDC2006T07. find
  59. Maamouri, M., Buckwalter, T., Graff, D., and Jin, H. (2007). Fisher levantine arabic conversational telephone speech, transcripts ldc2007t04. find
  60. McNeil, K. and Faiza, M. (2011). Tunisian Arabic Corpus : Creating a Written Corpus of an ”Unwritten” Language. In Proceedings of the Workshop on Arabic Corpus Linguistics (WACL). find
  61. Meftouh, K., Harrat, S., Jamoussi, S., Abbas, M., and Smaili, K. (2015). Machine translation experiments on PADIC: A parallel Arabic DIalect corpus. In Proceedings of the 29th Pacific Asia Conference on Language, Information and Computation, pages 26–34, Shanghai, China, October. find
  62. Mubarak, H. and Darwish, K. (2014). Using Twitter to collect a multi-dialectal corpus of Arabic. In Proceedings of the Workshop for Arabic Natural Language Processing (WANLP), Doha, Qatar. find
  63. Sadat, F., Kazemi, F., and Farzindar, A. (2014). Automatic Identification of Arabic Dialects in Social Media. In Proceedings of the Workshop on Natural Language Processing for Social Media (SocialNLP), pages 22–27, Dublin, Ireland. find
  64. Salama, A., Bouamor, H., Mohit, B., and Oflazer, K. (2014). YouDACC: the Youtube Dialectal Arabic Comment Corpus. In Proceedings of the Language Resources and Evaluation Conference (LREC), pages 1246–1251, Reykjavik, Iceland. find
  65. Sawalha, M., Alshargi, F., AlShdaifat, A., Yagi, S., and Qudah, M. A. (2019). Construction and annotation of the jordan comprehensive contemporary Arabic corpus (JCCA). In Proceedings of the Fourth Arabic Natural Language Processing Workshop, pages 148–157, Florence, Italy, August. Association for Computational Linguistics. find
  66. Shon, S., , Ali, A., Samih, Y., Mubarak, H., and Glass, J. (2020). Adi17: A fine-grained arabic dialect identification dataset. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 8244–8248. find our card
  67. Smaı̈li, K., Abbas, M., Meftouh, K., and Harrat, S. (2014). Building resources for Algerian Arabic dialects. In Proceedings of the Conference of the International Speech Communication Association (Interspeech). find
  68. Zaghouani, W. and Charfi, A. (2018). ArapTweet: A Large Multi-Dialect Twitter Corpus for Gender, Age and Language Variety Identification. In Proceedings of the Language Resources and Evaluation Conference (LREC), Miyazaki, Japan. find
  69. Zaidan, O. F. and Callison-Burch, C. (2011). The Arabic Online Commentary Dataset: an Annotated Dataset of Informal Arabic With High Dialectal Content. In Proceedings of the Conference of the Association for Computational Linguistics (ACL), pages 37–41. find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/baimukan-2022-hierarchical-aggregation-dialectal-data-arabic/