---
title: "Arabic Dialect Identification in the Wild — Abdelali 2020"
description: "Arabic speakers on social media mostly write in regional dialects, yet earlier dialect-identification resources were narrow in genre, unevenly spread…"
authors: ["Ahmed Abdelali", "Hamdy Mubarak", "Younes Samih", "Sabit Hassan", "Kareem Darwish"]
year: 2020
venue: "arXiv:2005.06557"
source: "https://arxiv.org/abs/2005.06557"
doi: "10.48550/arxiv.2005.06557"
arxiv: "2005.06557"
licence: "CC BY 4.0 (arXiv)"
licence_url: "http://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/abdelali-2020-arabic-dialect-identification-wild/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/abdelali-2020-arabic-dialect-identification-wild/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2020](https://md.research.elamur.ai/free/en/papers/2020/) / Arabic Dialect Identification in the Wild

# Arabic Dialect Identification in the Wild

**Authors:** [Ahmed Abdelali](https://md.research.elamur.ai/free/en/authors/ahmed-abdelali/), [Hamdy Mubarak](https://md.research.elamur.ai/free/en/authors/hamdy-mubarak/), [Younes Samih](https://md.research.elamur.ai/free/en/authors/younes-samih/), [Sabit Hassan](https://md.research.elamur.ai/free/en/authors/sabit-hassan/), [Kareem Darwish](https://md.research.elamur.ai/free/en/authors/kareem-darwish/) · **Year:** 2020 · **Venue:** arXiv:2005.06557 · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2005.06557) · [PDF](https://arxiv.org/pdf/2005.06557) · [DOI 10.48550/arxiv.2005.06557](https://doi.org/10.48550/arxiv.2005.06557) · [arXiv 2005.06557](https://arxiv.org/abs/2005.06557) · [OpenAlex](https://openalex.org/W3025939269)

**Citations:** 24 citations (openalex-snapshot · 2026-10-14) · [22](#cited-by) in our library, of the 140884 papers whose bibliographies we read

## Summary
Arabic speakers on social media mostly write in regional dialects, yet earlier dialect-identification resources were narrow in genre, unevenly spread across countries, or dominated by Modern Standard Arabic (MSA). The authors present QADI, a balanced, country-level corpus of dialectal tweets collected without manual labeling. They started from a 25M-tweet Arabic stream and ran a filter cascade. First, a gazetteer of country names, city names and nationality adjectives matched users whose profiles state where they are from. Next, a fastText classifier separated dialect from MSA, and only accounts that mostly write in dialect were kept. Finally, an obscenity lexicon removed accounts that post mainly vulgar content. The resulting corpus has 540k tweets from 2,525 accounts across 18 countries. A manual check estimated the labels at 91.5% accuracy. A fine-tuned AraBERT model reached 60.6% macro-F1 over 18 classes. The same models trained on MADAR tweets scored far lower, because most of that data is MSA. The collection method can be rerun at any time to gather fresh data.

## Key points
- QADI has 540k tweets from 2,525 users, averaging about 140 accounts per country, across 18 Arab countries. Mauritania, Somalia, Djibouti and Comoros were left out because too few users could be found. The test set has about 182 tweets per country, labeled by native speakers.
- The MSA-vs-dialect filter was trained on 3.09M MSA and 3.17M dialectal tweets. These were labeled heuristically by relative pronoun: the MSA forms versus the dialectal "Ally". Using character 3–6-grams, fastText reached 98% accuracy on a held-out 20k set.
- Users were retained only when dialect accounted for 50% or more of their tweets. They were dropped if more than 50% contained words from an obscenity list of 288 words and 127 hashtags.
- Macro-F1 when training on QADI versus training on MADAR:

  | Model | QADI | MADAR |
  |---|---|---|
  | AraBERT | 60.6 | 29.0 |
  | MultiLangBERT | 58.9 | 25.3 |
  | SVM (CW26) | 57.2 | 24.0 |
  | SVM (C37) | 48.1 | 21.2 |
  | Mazajak | 39.8 | 24.6 |

  The dialect filter classified 71% of MADAR tweets as MSA, which explains the gap.
- Most errors fall inside a region, especially among the Gulf states, LB vs SY, MA vs DZ, and EG vs SD. Only about 2% of errors land outside the correct region, often in tweets that quote songs or poems or contain MSA words. Scores also vary strongly with tweet length, from roughly 46 to 73 F1.

## Interesting ideas
- The authors labeled millions of examples using one marker word that reliably signals the class (the dialectal relative pronoun), then masked that word so the classifier had to learn other cues. — For Mova (learn): Mova could build silver data from unambiguous trigger constructions, then hide the trigger during training and keep the unmasked version as a negative control. This would check that the model generalizes instead of memorizing the trigger.
- Users' own statements of origin served as free weak labels, and a gazetteer turned this metadata into country tags without any annotation. — For Mova (tree): Source metadata, such as a publisher's spelling variety or region, could cheaply tag English sentences by dialect. Those tags could then select the right UD conversion or normalization rules per treebank.
- Filtering was done at the account level with simple majority thresholds, so whole noisy sources were removed rather than individual items. — For Mova (learn): When agreeing silver parses across witness parsers, Mova could drop entire documents or sources whose share of disagreements passes a threshold. This keeps systematically bad sources out of training.
- Platform-specific tokens were replaced with placeholders (@USER, NUM, EMOJI, URL) and hashtags were split into words, which removed non-linguistic shortcuts. — For Mova (tokens): The en tokenizer could map web artifacts to typed placeholders and segment hashtags before tagging. The parser would then see clean, explainable token classes.
- The error analysis traced out-of-region mistakes to embedded quotations from songs or poems, whose language differs from the writer's own. — For Mova (frames): Detecting quoted or recited material separately would stop prag from attributing a quote's speech act or claims to the narrator. Mova could also report scores by input-length bins, as the paper does.

*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](https://md.research.elamur.ai/free/en/shelves/linguistics/)

## Cited by

Cited by

22 papers of our library cite it (of the 140884 whose bibliographies we read); 8 of them have a page here, 14 not yet.

- [Arabic Dialect Classification using RNNs, Transformers, and Large Language Models: A Comparative Analysis](https://md.research.elamur.ai/free/en/papers/a-2025-essameldin-arabic-dialect-classification-using-rnns/)
- [Pre-Training BERT on Arabic Tweets: Practical Considerations](https://md.research.elamur.ai/free/en/papers/abdelali-2021-pre-training-bert-arabic-tweets/)
- [ARBERT & MARBERT: Deep Bidirectional Transformers for Arabic](https://md.research.elamur.ai/free/en/papers/abdul-2021-mageed-arbert-marbert-deep-bidirectional-transformers/)
- [Hierarchical Aggregation of Dialectal Data for Arabic Dialect Identification](https://md.research.elamur.ai/free/en/papers/baimukan-2022-hierarchical-aggregation-dialectal-data-arabic/)
- [Arabic Diacritization Using Morphologically Informed Character-Level Model](https://md.research.elamur.ai/free/en/papers/coling-2024-elmallah-arabic-diacritization-morphologically-informed-character/)
- [The Interplay of Variant, Size, and Task Type in Arabic Pre-trained Language Models](https://md.research.elamur.ai/free/en/papers/inoue-2021-interplay-variant-size-task-type/)
- [ArCovidVac: Analyzing Arabic Tweets About COVID-19 Vaccination](https://md.research.elamur.ai/free/en/papers/mubarak-2022-arcovidvac-analyzing-arabic-tweets-about/)
- [AraT5: Text-to-Text Transformers for Arabic Language Generation](https://md.research.elamur.ai/free/en/papers/nagoudi-2022-arat5-text-text-transformers-arabic/)

## Cite

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

```bibtex
@misc{abdelali2020arabic,
  title = {{Arabic Dialect Identification in the Wild}},
  author = {Abdelali, Ahmed and Mubarak, Hamdy and Samih, Younes and Hassan, Sabit and Darwish, Kareem},
  year = {2020},
  doi = {10.48550/arxiv.2005.06557},
  eprint = {2005.06557},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2005.06557}
}
```

- [← previous Arabic Diacritic Recovery Using a Feature-Rich biLSTM Model](https://md.research.elamur.ai/free/en/papers/darwish-2020-arabic-diacritic-recovery-using-feature/)
- [next → Arabic Dialects of Lamomea and Pudahoa at Students in Gontor 4 and 7 Southeast Sulawesi](https://md.research.elamur.ai/free/en/papers/wahab-2020-arabic-dialects-lamomea-pudahoa-students/)
- [↑ back to papers of 2020](https://md.research.elamur.ai/free/en/papers/2020/)

## References

35 references, 7 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Muhammad Abdul-Mageed, Hassan Alhuzali, and Mohamed Elaraby. 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](https://scholar.google.com/scholar?q=Muhammad+Abdul%2DMageed%2C+Hassan+Alhuzali%2C+and+Mohamed+Elaraby%2E+2018%2E+You+tweet+what+you+speak%3A+A+city%2Dlevel+dataset+of+Arabic+dialects%2E+In+Proceedings+of+the+Language+Resources+and+Evaluation+Conference)
2. Ibrahim Abu Farha and Walid Magdy. 2019. Mazajak: An online Arabic sentiment analyser. In Proceedings of the Fourth Arabic Natural Language Processing Workshop, pages 192–198, Florence, Italy, August. Association for Computational Linguistics. [https://en.wikipedia.org/wiki/Egyptian_Arabic](https://en.wikipedia.org/wiki/Egyptian_Arabic) [↗](https://en.wikipedia.org/wiki/Egyptian_Arabic)
3. Areej Alshutayri and Erik Atwell. 2017. Exploring twitter as a source of an arabic dialect corpus. International Journal of Computational Linguistics (IJCL), 8(2):37–44, June. This is an open access article under the terms of the Creative Commons Attribution License (CC-BY). [find](https://scholar.google.com/scholar?q=Areej+Alshutayri+and+Erik+Atwell%2E+2017%2E+Exploring+twitter+as+a+source+of+an+arabic+dialect+corpus%2E+International+Journal+of+Computational+Linguistics+%28IJCL%29%2C+8%282%29%3A37%E2%80%9344%2C+June%2E+This+is+an+open+access+a)
4. Wissam Antoun, Fady Baly, and Hazem Hajj. 2020. Arabert: Transformer-based model for arabic language understanding. In Proceedings of The 4th Workshop on Open-Source Arabic Corpora and Processing Tools, pages 9–15. [find](https://scholar.google.com/scholar?q=Wissam+Antoun%2C+Fady+Baly%2C+and+Hazem+Hajj%2E+2020%2E+Arabert%3A+Transformer%2Dbased+model+for+arabic+language+understanding%2E+In+Proceedings+of+The+4th+Workshop+on+Open%2DSource+Arabic+Corpora+and+Processing+Tool) [our card](https://md.research.elamur.ai/free/en/papers/antoun-2020-arabert-transformer-based-model-arabic/)
5. Piyush Bansal, Romil Bansal, and Vasudeva Varma. 2015. Towards deep semantic analysis of hashtags. In European conference on information retrieval, pages 453–464. Springer. [find](https://scholar.google.com/scholar?q=Piyush+Bansal%2C+Romil+Bansal%2C+and+Vasudeva+Varma%2E+2015%2E+Towards+deep+semantic+analysis+of+hashtags%2E+In+European+conference+on+information+retrieval%2C+pages+453%E2%80%93464%2E+Springer%2E)
6. Houda Bouamor, Nizar Habash, Mohammad Salameh, Wajdi Zaghouani, Owen Rambow, Dana Abdulrahim, Ossama Obeid, Salam Khalifa, Fadhl Eryani, Alexander Erdmann, and Kemal Oflazer. 2018. The MADAR Arabic Dialect Corpus and Lexicon. In Proceedings of the Language Resources and Evaluation Conference (LREC), Miyazaki, Japan. [find](https://scholar.google.com/scholar?q=Houda+Bouamor%2C+Nizar+Habash%2C+Mohammad+Salameh%2C+Wajdi+Zaghouani%2C+Owen+Rambow%2C+Dana+Abdulrahim%2C+Ossama+Obeid%2C+Salam+Khalifa%2C+Fadhl+Eryani%2C+Alexander+Erdmann%2C+and+Kemal+Oflazer%2E+2018%2E+The+MADAR+Arabic+Di)
7. Houda Bouamor, Sabit Hassan, and Nizar Habash. 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](https://scholar.google.com/scholar?q=Houda+Bouamor%2C+Sabit+Hassan%2C+and+Nizar+Habash%2E+2019%2E+The+MADAR+shared+task+on+Arabic+fine%2Dgrained+dialect+identification%2E+In+Proceedings+of+the+Fourth+Arabic+Natural+Language+Processing+Workshop%2C+page) [our card](https://md.research.elamur.ai/free/en/papers/bouamor-2019-madar-shared-task-arabic-fine/)
8. Michael Conover, Jacob Ratkiewicz, Matthew R Francisco, Bruno Gonçalves, Filippo Menczer, and Alessandro Flammini. 2011. Political polarization on twitter. ICWSM, 133:89–96. [find](https://scholar.google.com/scholar?q=Michael+Conover%2C+Jacob+Ratkiewicz%2C+Matthew+R+Francisco%2C+Bruno+Gonc%CC%A7alves%2C+Filippo+Menczer%2C+and+Alessandro+Flammini%2E+2011%2E+Political+polarization+on+twitter%2E+ICWSM%2C+133%3A89%E2%80%9396%2E)
9. Kareem Darwish and Hamdy Mubarak. 2016. Farasa: A new fast and accurate arabic word segmenter. In Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC’16), pages 1070–1074. [find](https://scholar.google.com/scholar?q=Kareem+Darwish+and+Hamdy+Mubarak%2E+2016%2E+Farasa%3A+A+new+fast+and+accurate+arabic+word+segmenter%2E+In+Proceedings+of+the+Tenth+International+Conference+on+Language+Resources+and+Evaluation+%28LREC%E2%80%9916%29%2C+page)
10. Kareem Darwish, Hassan Sajjad, and Hamdy Mubarak. 2014. Verifiably Effective Arabic Dialect Identification. In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha, Qatar. [find](https://scholar.google.com/scholar?q=Kareem+Darwish%2C+Hassan+Sajjad%2C+and+Hamdy+Mubarak%2E+2014%2E+Verifiably+Effective+Arabic+Dialect+Identification%2E+In+Proceedings+of+the+Conference+on+Empirical+Methods+in+Natural+Language+Processing+%28EMNLP%29)
11. Kareem Darwish, Mohammed Attia, Hamdy Mubarak, Younes Samih, Ahmed Abdelali, Lluı́s Màrquez, Mohamed Eldesouki, and Laura Kallmeyer. 2020. Effective multi dialectal arabic pos tagging. Natural Language Engineering, 1(1):18. [find](https://scholar.google.com/scholar?q=Kareem+Darwish%2C+Mohammed+Attia%2C+Hamdy+Mubarak%2C+Younes+Samih%2C+Ahmed+Abdelali%2C+Llu%C4%B1%CC%81s+Ma%CC%80rquez%2C+Mohamed+Eldesouki%2C+and+Laura+Kallmeyer%2E+2020%2E+Effective+multi+dialectal+arabic+pos+tagging%2E+Natural+Langua)
12. Thierry Declerck and Piroska Lendvai. 2015. Processing and normalizing hashtags. In Proceedings of the International Conference Recent Advances in Natural Language Processing, pages 104–109. [find](https://scholar.google.com/scholar?q=Thierry+Declerck+and+Piroska+Lendvai%2E+2015%2E+Processing+and+normalizing+hashtags%2E+In+Proceedings+of+the+International+Conference+Recent+Advances+in+Natural+Language+Processing%2C+pages+104%E2%80%93109%2E)
13. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language understanding. In 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), pages 4171–4186, Minneapolis, Minnesota, June. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Jacob+Devlin%2C+Ming%2DWei+Chang%2C+Kenton+Lee%2C+and+Kristina+Toutanova%2E+2019%2E+BERT%3A+Pre%2Dtraining+of+deep+bidirectional+transformers+for+language+understanding%2E+In+Proceedings+of+the+2019+Conference+of+the+N)
14. Mona T Diab, Mohamed Al-Badrashiny, Maryam Aminian, Mohammed Attia, Heba Elfardy, Nizar Habash, Abdelati Hawwari, Wael Salloum, Pradeep Dasigi, and Ramy Eskander. 2014. Tharwa: A Large Scale Dialectal Arabic-Standard Arabic-English Lexicon. In Proceedings of the Language Resources and Evaluation Conference (LREC), pages 3782–3789, Reykjavik, Iceland. [find](https://scholar.google.com/scholar?q=Mona+T+Diab%2C+Mohamed+Al%2DBadrashiny%2C+Maryam+Aminian%2C+Mohammed+Attia%2C+Heba+Elfardy%2C+Nizar+Habash%2C+Abdelati+Hawwari%2C+Wael+Salloum%2C+Pradeep+Dasigi%2C+and+Ramy+Eskander%2E+2014%2E+Tharwa%3A+A+Large+Scale+Dialectal)
15. Mahmoud El-Haj, Paul Rayson, and Mariam Aboelezz. 2018. Arabic dialect identification in the context of bivalency and code-switching. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018), Miyazaki, Japan, May. European Language Resources Association (ELRA). [find](https://scholar.google.com/scholar?q=Mahmoud+El%2DHaj%2C+Paul+Rayson%2C+and+Mariam+Aboelezz%2E+2018%2E+Arabic+dialect+identification+in+the+context+of+bivalency+and+code%2Dswitching%2E+In+Proceedings+of+the+Eleventh+International+Conference+on+Languag)
16. Heba Elfardy and Mona Diab. 2013. Sentence Level Dialect Identification in Arabic. In Proceedings of the Conference of the Association for Computational Linguistics (ACL), Sofia, Bulgaria. [find](https://scholar.google.com/scholar?q=Heba+Elfardy+and+Mona+Diab%2E+2013%2E+Sentence+Level+Dialect+Identification+in+Arabic%2E+In+Proceedings+of+the+Conference+of+the+Association+for+Computational+Linguistics+%28ACL%29%2C+Sofia%2C+Bulgaria%2E)
17. Jeremy Howard and Sebastian Ruder. 2018. Universal language model fine-tuning for text classification. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Melbourne, Australia, July. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Jeremy+Howard+and+Sebastian+Ruder%2E+2018%2E+Universal+language+model+fine%2Dtuning+for+text+classification%2E+In+Proceedings+of+the+56th+Annual+Meeting+of+the+Association+for+Computational+Linguistics+%28Volum)
18. Radu Tudor Ionescu and Marius Popescu. 2016. UnibucKernel: An approach for Arabic dialect identification based on multiple string kernels. In Proceedings of the Third Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial3), pages 135–144, Osaka, Japan, December. [find](https://scholar.google.com/scholar?q=Radu+Tudor+Ionescu+and+Marius+Popescu%2E+2016%2E+UnibucKernel%3A+An+approach+for+Arabic+dialect+identification+based+on+multiple+string+kernels%2E+In+Proceedings+of+the+Third+Workshop+on+NLP+for+Similar+Langu)
19. The COLING 2016 Organizing Committee. [find](https://scholar.google.com/scholar?q=The+COLING+2016+Organizing+Committee%2E)
20. Armand Joulin, Edouard Grave, Piotr Bojanowski, and Tomas Mikolov. 2016. Bag of tricks for efficient text classification. arXiv preprint arXiv:1607.01759. [↗](https://arxiv.org/abs/1607.01759)
21. X. Li and J. Huang. 2009. Shc: A spectral algorithm for hierarchical clustering. In 2009 International Conference on Multimedia Information Networking and Security, volume 2, pages 197–200. [find](https://scholar.google.com/scholar?q=X%2E+Li+and+J%2E+Huang%2E+2009%2E+Shc%3A+A+spectral+algorithm+for+hierarchical+clustering%2E+In+2009+International+Conference+on+Multimedia+Information+Networking+and+Security%2C+volume+2%2C+pages+197%E2%80%93200%2E)
22. Laurens van der Maaten and Geoffrey Hinton. 2008. Visualizing data using t-sne. Journal of machine learning research, 9(Nov):2579–2605. [find](https://scholar.google.com/scholar?q=Laurens+van+der+Maaten+and+Geoffrey+Hinton%2E+2008%2E+Visualizing+data+using+t%2Dsne%2E+Journal+of+machine+learning+research%2C+9%28Nov%29%3A2579%E2%80%932605%2E)
23. Shervin Malmasi, Marcos Zampieri, Nikola Ljubešić, Preslav Nakov, Ahmed Ali, and Jörg Tiedemann. 2016. Discriminating between similar languages and Arabic dialect identification: A report on the third DSL shared task. In Proceedings of the Third Workshop on NLP for Similar Languages, Varieties and Dialects (VarDial3), pages 1–14, Osaka, Japan, December. [find](https://scholar.google.com/scholar?q=Shervin+Malmasi%2C+Marcos+Zampieri%2C+Nikola+Ljubes%CC%8Cic%CC%81%2C+Preslav+Nakov%2C+Ahmed+Ali%2C+and+Jo%CC%88rg+Tiedemann%2E+2016%2E+Discriminating+between+similar+languages+and+Arabic+dialect+identification%3A+A+report+on+the+th) [our card](https://md.research.elamur.ai/free/en/papers/malmasi-2016-discriminating-between-similar-languages/)
24. The COLING 2016 Organizing Committee. [find](https://scholar.google.com/scholar?q=The+COLING+2016+Organizing+Committee%2E)
25. Hamdy Mubarak and Kareem Darwish. 2014. Using twitter to collect a multi-dialectal corpus of arabic. In Proceedings of the EMNLP 2014 Workshop on Arabic Natural Language Processing (ANLP), pages 1–7. [find](https://scholar.google.com/scholar?q=Hamdy+Mubarak+and+Kareem+Darwish%2E+2014%2E+Using+twitter+to+collect+a+multi%2Ddialectal+corpus+of+arabic%2E+In+Proceedings+of+the+EMNLP+2014+Workshop+on+Arabic+Natural+Language+Processing+%28ANLP%29%2C+pages+1%E2%80%937%2E)
26. Hamdy Mubarak, Kareem Darwish, and Walid Magdy. 2017. Abusive language detection on arabic social media. In Proceedings of the First Workshop on Abusive Language Online, pages 52–56. [find](https://scholar.google.com/scholar?q=Hamdy+Mubarak%2C+Kareem+Darwish%2C+and+Walid+Magdy%2E+2017%2E+Abusive+language+detection+on+arabic+social+media%2E+In+Proceedings+of+the+First+Workshop+on+Abusive+Language+Online%2C+pages+52%E2%80%9356%2E) [our card](https://md.research.elamur.ai/free/en/papers/mubarak-2017-abusive-language-detection-arabic-social/)
27. Damian Radcliffe and Payton Bruni. 2019. State of Social Media Middle East: 2018. University of Oregon Libraries. [find](https://scholar.google.com/scholar?q=Damian+Radcliffe+and+Payton+Bruni%2E+2019%2E+State+of+Social+Media+Middle+East%3A+2018%2E+University+of+Oregon+Libraries%2E)
28. Alec Radford, Karthik Narasimhan, Tim Salimans, and Ilya Sutskever. 2018. language understanding by generative pre-training. URL [https://s3-us-west-2.com/openai-assets/researchcovers/languageunsupervised/language](https://s3-us-west-2.com/openai-assets/researchcovers/languageunsupervised/language) understanding paper. pdf. Improving amazonaws. [↗](https://s3-us-west-2.com/openai-assets/researchcovers/languageunsupervised/language)
29. Fatiha Sadat, Farnazeh Kazemi, and Atefeh Farzindar. 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](https://scholar.google.com/scholar?q=Fatiha+Sadat%2C+Farnazeh+Kazemi%2C+and+Atefeh+Farzindar%2E+2014%2E+Automatic+Identification+of+Arabic+Dialects+in+Social+Media%2E+In+Proceedings+of+the+Workshop+on+Natural+Language+Processing+for+Social+Media+%28)
30. Mohammad Salameh, Houda Bouamor, and Nizar Habash. 2018. Fine-grained arabic dialect identification. In Proceedings of the 27th International Conference on Computational Linguistics, Santa Fe, New Mexico, USA. [find](https://scholar.google.com/scholar?q=Mohammad+Salameh%2C+Houda+Bouamor%2C+and+Nizar+Habash%2E+2018%2E+Fine%2Dgrained+arabic+dialect+identification%2E+In+Proceedings+of+the+27th+International+Conference+on+Computational+Linguistics%2C+Santa+Fe%2C+New+Mex) [our card](https://md.research.elamur.ai/free/en/papers/salameh-2018-fine-grained-arabic-dialect-identification/)
31. Younes Samih, Mohamed Eldesouki, Mohammed Attia, Kareem Darwish, Ahmed Abdelali, Hamdy Mubarak, and Laura Kallmeyer. 2017. Learning from relatives: Unified dialectal Arabic segmentation. In Proceedings of the 21st Conference on Computational Natural Language Learning (CoNLL 2017), pages 432–441, Vancouver, Canada, August. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Younes+Samih%2C+Mohamed+Eldesouki%2C+Mohammed+Attia%2C+Kareem+Darwish%2C+Ahmed+Abdelali%2C+Hamdy+Mubarak%2C+and+Laura+Kallmeyer%2E+2017%2E+Learning+from+relatives%3A+Unified+dialectal+Arabic+segmentation%2E+In+Proceeding)
32. Wajdi Zaghouani and Anis Charfi. 2018. Arap-tweet: A large multi-dialect twitter corpus for gender, age and language variety identification. CoRR, abs/1808.07674. [↗](https://arxiv.org/abs/1808.07674)
33. Omar F Zaidan and Chris Callison-Burch. 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](https://scholar.google.com/scholar?q=Omar+F+Zaidan+and+Chris+Callison%2DBurch%2E+2011%2E+The+Arabic+Online+Commentary+Dataset%3A+an+Annotated+Dataset+of+Informal+Arabic+With+High+Dialectal+Content%2E+In+Proceedings+of+the+Conference+of+the+Associa) [our card](https://md.research.elamur.ai/free/en/papers/zaidan-2011-arabic-online-commentary-dataset-annotated/)
34. Omar F Zaidan and Chris Callison-Burch. 2014. Arabic dialect identification. Computational Linguistics, 40(1):171–202. [find](https://scholar.google.com/scholar?q=Omar+F+Zaidan+and+Chris+Callison%2DBurch%2E+2014%2E+Arabic+dialect+identification%2E+Computational+Linguistics%2C+40%281%29%3A171%E2%80%93202%2E) [our card](https://md.research.elamur.ai/free/en/papers/zaidan-2014-arabic-dialect-identification/)
35. Chiyu Zhang and Muhammad Abdul-Mageed. 2019. No army, no navy: Bert semi-supervised learning of arabic dialects. pages 279–284, 01. [find](https://scholar.google.com/scholar?q=Chiyu+Zhang+and+Muhammad+Abdul%2DMageed%2E+2019%2E+No+army%2C+no+navy%3A+Bert+semi%2Dsupervised+learning+of+arabic+dialects%2E+pages+279%E2%80%93284%2C+01%2E)