---
title: "ANER: Arabic and Arabizi Named Entity — Sadallah 2023"
description: "Named entity recognition for Arabic is hard for several reasons."
authors: ["Abdelrahman \"Boda\" Sadallah", "Omar Ahmed", "Shimaa Mohamed", "Omar Hatem", "Doaa Hesham", "Ahmed H. Yousef"]
year: 2023
venue: "arXiv:2308.14669"
source: "https://arxiv.org/abs/2308.14669"
doi: "10.1109/imsa58542.2023.10217635"
arxiv: "2308.14669"
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/sadallah-2023-aner-arabic-arabizi-named-entity/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/sadallah-2023-aner-arabic-arabizi-named-entity/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2023](https://md.research.elamur.ai/free/en/papers/2023/) / ANER: Arabic and Arabizi Named Entity Recognition using Transformer-Based Approach

# ANER: Arabic and Arabizi Named Entity Recognition using Transformer-Based Approach

**Authors:** Abdelrahman "Boda" Sadallah, Omar Ahmed, Shimaa Mohamed, Omar Hatem, Doaa Hesham, Ahmed H. Yousef · **Year:** 2023 · **Venue:** arXiv:2308.14669 · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://doi.org/10.1109/imsa58542.2023.10217635) · [PDF](https://arxiv.org/pdf/2308.14669) · [DOI 10.1109/imsa58542.2023.10217635](https://doi.org/10.1109/imsa58542.2023.10217635) · [arXiv 2308.14669](https://arxiv.org/abs/2308.14669) · [OpenAlex](https://openalex.org/W4386126687)

**Citations:** 3 citations (openalex-snapshot · 2026-10-14)

## Summary
Named entity recognition for Arabic is hard for several reasons. The morphology is dense with clitics. Diacritics are usually omitted, which makes words ambiguous. There is no capitalization, and dialects vary widely. Existing tools also mostly stop at four coarse classes. The authors built ANER, a fine-grained recognizer that fine-tunes AraBERTv0.2-base with a token-classification head. They trained it on WikiFANE Gold, a manually annotated Wikipedia corpus covering 50 entity types. For an out-of-domain check they used NewsFANE Gold, which takes the news text of ANERcorp and relabels it with the same 50-type scheme. A separate instance was trained on ANERcorp using the CAMeL Tools splits.

The reported F1 is 88.7%, set against the 83% that CAMeL Tools gets on the four-class ANERcorp. On out-of-domain news the F1 is 77.7%. The system is packaged as a public web application. It accepts Arabic, Arabizi (Arabic written in Latin letters and digits) or a mix of both. Recognized entities are highlighted and link to Wikipedia. The model is also published on HuggingFace, and the training scripts are open-sourced.

## Key points
- The base model is AraBERTv0.2-base: 12 encoder layers, 12 heads, hidden size 768, about 110M parameters. A fully connected layer with 102 outputs (101 entity labels plus non-entity) and cross-entropy loss sits on top.
- The data comes from King Abdulaziz University:
  - WikiFANE Gold has about 500k tokens and 50 classes, split 80/10/10.
  - NewsFANE Gold has about 170k tokens and serves as the out-of-distribution test.
  - ANERcorp has about 150k tokens from 316 newswire articles, in CoNLL format with 4 classes.
- Training used a sequence length of 256 (chosen after checking sentence lengths), 10 epochs and a learning rate of 5×10⁻⁴, on free Google Colab with 15GB of GPU memory.
- Results: 88.7% F1 versus CAMeL Tools' 83% on ANERcorp, and 77.7% F1 on NewsFANE Gold.
- Deployment:
  - Website: aner.online, where users can pick either the ANER model or the CAMeL Tools model.
  - HuggingFace model: boda/ANER.
  - Arabizi support depends on how well the transliteration step works.

## Interesting ideas
- Because NewsFANE Gold annotates the same text as ANERcorp under a richer label scheme, the effect of label granularity can be separated from the effect of domain. — For Mova (learn): Mova could annotate one fixed text set under both coarse and fine frame or role inventories, so evaluation shows whether errors come from finer distinctions or from new text.
- The authors labeled subword pieces in two ways, either every piece of a word or only the first piece, and found little difference between them. — For Mova (tree): gparse could run the same quick ablation on how word-level heads and labels map onto subword pieces, and settle on the simpler scheme if the scores match.
- Output entities are made clickable and lead straight to an external knowledge page (Wikipedia), so each prediction can be checked by the user. — For Mova (world): entities in the world of a text could carry a link to their Wikidata or WordNet node, which makes the grounding step inspectable.
- Nonstandard script input (Arabizi) is converted to the standard form before the model sees it, and the authors note that transliteration errors pass straight into recognition errors. — For Mova (tokens): a separate normalization pass for informal English spellings could sit ahead of the tagger, with its errors logged on their own so they are not blamed on the tagger.
- The interface lets users run their own model or a baseline model on the same input side by side, which makes qualitative comparison immediate. — For Mova (learn): a viewer that puts the parses or frames of witness parsers next to each other on one sentence would help when deciding on silver-data agreement.

*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/)

Topics: [Named entity recognition](https://md.research.elamur.ai/free/en/topics/named-entity-recognition/)

## Cite

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

```bibtex
@misc{sadallah2023aner,
  title = {{ANER: Arabic and Arabizi Named Entity Recognition using Transformer-Based Approach}},
  author = {Sadallah, Abdelrahman "Boda" and Ahmed, Omar and Mohamed, Shimaa and Hatem, Omar and Hesham, Doaa and Yousef, Ahmed H.},
  year = {2023},
  doi = {10.1109/imsa58542.2023.10217635},
  eprint = {2308.14669},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.1109/imsa58542.2023.10217635}
}
```

- [← previous Analyzing Transformers in Embedding Space](https://md.research.elamur.ai/free/en/papers/dar-2023-analyzing-transformers-embedding-space/)
- [next → AnnoLLM: Making Large Language Models to Be Better Crowdsourced Annotators](https://md.research.elamur.ai/free/en/papers/he-2023-annollm-making-large-language-models/)
- [↑ back to papers of 2023](https://md.research.elamur.ai/free/en/papers/2023/)

## References

17 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. E. F. Tjong Kim Sang and F. De Meulder, “Introduction to the CoNLL-2003 shared task: Language-independent named entity recognition,” in Proceedings of the Seventh Conference on Natural Language Learning at HLT-NAACL 2003, 2003, pp. 142–147. [find](https://scholar.google.com/scholar?q=E%2E+F%2E+Tjong+Kim+Sang+and+F%2E+De+Meulder%2C+%E2%80%9CIntroduction+to+the+CoNLL%2D2003+shared+task%3A+Language%2Dindependent+named+entity+recognition%2C%E2%80%9D+in+Proceedings+of+the+Seventh+Conference+on+Natural+Language+Learni)
2. W. Antoun, F. Baly, and H. Hajj, “AraBERT: Transformer-based model for Arabic language understanding,” in Proceedings of the 4th Workshop on Open-Source Arabic Corpora and Processing Tools, with a Shared Task on Offensive Language Detection. Marseille, France: European Language Resource Association, May 2020, pp. 9–15. [find](https://scholar.google.com/scholar?q=W%2E+Antoun%2C+F%2E+Baly%2C+and+H%2E+Hajj%2C+%E2%80%9CAraBERT%3A+Transformer%2Dbased+model+for+Arabic+language+understanding%2C%E2%80%9D+in+Proceedings+of+the+4th+Workshop+on+Open%2DSource+Arabic+Corpora+and+Processing+Tools%2C+with+a+Sha) [our card](https://md.research.elamur.ai/free/en/papers/antoun-2020-arabert-transformer-based-model-arabic/)
3. J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “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). Minneapolis, Minnesota: Association for Computational Linguistics, Jun. 2019, pp. 4171–4186. [find](https://scholar.google.com/scholar?q=J%2E+Devlin%2C+M%2E%2DW%2E+Chang%2C+K%2E+Lee%2C+and+K%2E+Toutanova%2C+%E2%80%9CBERT%3A+Pre%2Dtraining+of+deep+bidirectional+transformers+for+language+understanding%2C%E2%80%9D+in+Proceedings+of+the+2019+Conference+of+the+North+American+Chapte)
4. A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. u. Kaiser, and I. Polosukhin, “Attention is all you need,” in Advances in Neural Information Processing Systems, I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett, Eds., vol. 30. Curran Associates, Inc., 2017. [find](https://scholar.google.com/scholar?q=A%2E+Vaswani%2C+N%2E+Shazeer%2C+N%2E+Parmar%2C+J%2E+Uszkoreit%2C+L%2E+Jones%2C+A%2E+N%2E+Gomez%2C+L%2E+u%2E+Kaiser%2C+and+I%2E+Polosukhin%2C+%E2%80%9CAttention+is+all+you+need%2C%E2%80%9D+in+Advances+in+Neural+Information+Processing+Systems%2C+I%2E+Guyon%2C+U%2E)
5. F. Alotaibi and M. Lee, “A hybrid approach to features representation for fine-grained Arabic named entity recognition,” in Proceedings of COLING 2014, the 25th International Conference on Computational Linguistics: Technical Papers. Dublin, Ireland: Dublin City University and Association for Computational Linguistics, Aug. 2014, pp. 984–995. [find](https://scholar.google.com/scholar?q=F%2E+Alotaibi+and+M%2E+Lee%2C+%E2%80%9CA+hybrid+approach+to+features+representation+for+fine%2Dgrained+Arabic+named+entity+recognition%2C%E2%80%9D+in+Proceedings+of+COLING+2014%2C+the+25th+International+Conference+on+Computation)
6. M. Maamouri, D. Graff, B. Bouziri, S. Krouna, A. Bies, and S. Kulick, “Ldc standard arabic morphological analyzer (sama) version 3.1,” Linguistic Data Consortium, 2010. [find](https://scholar.google.com/scholar?q=M%2E+Maamouri%2C+D%2E+Graff%2C+B%2E+Bouziri%2C+S%2E+Krouna%2C+A%2E+Bies%2C+and+S%2E+Kulick%2C+%E2%80%9CLdc+standard+arabic+morphological+analyzer+%28sama%29+version+3%2E1%2C%E2%80%9D+Linguistic+Data+Consortium%2C+2010%2E)
7. N. Habash, R. Eskander, and A. Hawwari, “A morphological analyzer for Egyptian Arabic,” in Proceedings of the Twelfth Meeting of the Special Interest Group on Computational Morphology and Phonology. Montréal, Canada: Association for Computational Linguistics, Jun. 2012, pp. 1–9. [find](https://scholar.google.com/scholar?q=N%2E+Habash%2C+R%2E+Eskander%2C+and+A%2E+Hawwari%2C+%E2%80%9CA+morphological+analyzer+for+Egyptian+Arabic%2C%E2%80%9D+in+Proceedings+of+the+Twelfth+Meeting+of+the+Special+Interest+Group+on+Computational+Morphology+and+Phonology%2E+M) [our card](https://md.research.elamur.ai/free/en/papers/lingbulk-2012-habash-morphological-analyzer-egyptian-arabic/)
8. I. A. Al-Sughaiyer and I. A. Al-Kharashi, “Arabic morphological analysis techniques: A comprehensive survey,” Journal of the American Society for Information Science and Technology, vol. 55, no. 3, pp. 189– 213, 2004. [find](https://scholar.google.com/scholar?q=I%2E+A%2E+Al%2DSughaiyer+and+I%2E+A%2E+Al%2DKharashi%2C+%E2%80%9CArabic+morphological+analysis+techniques%3A+A+comprehensive+survey%2C%E2%80%9D+Journal+of+the+American+Society+for+Information+Science+and+Technology%2C+vol%2E+55%2C+no%2E+3%2C+pp)
9. O. Obeid, N. Zalmout, S. Khalifa, D. Taji, M. Oudah, B. Alhafni, G. Inoue, F. Eryani, A. Erdmann, and N. Habash, “CAMeL tools: An open source python toolkit for Arabic natural language processing,” in Proceedings of the Twelfth Language Resources and Evaluation Conference. Marseille, France: European Language Resources Association, May 2020, pp. 7022–7032. [find](https://scholar.google.com/scholar?q=O%2E+Obeid%2C+N%2E+Zalmout%2C+S%2E+Khalifa%2C+D%2E+Taji%2C+M%2E+Oudah%2C+B%2E+Alhafni%2C+G%2E+Inoue%2C+F%2E+Eryani%2C+A%2E+Erdmann%2C+and+N%2E+Habash%2C+%E2%80%9CCAMeL+tools%3A+An+open+source+python+toolkit+for+Arabic+natural+language+processing%2C%E2%80%9D+in) [our card](https://md.research.elamur.ai/free/en/papers/obeid-2020-camel-tools-open-source-python/)
10. A. Abdelali, K. Darwish, N. Durrani, and H. Mubarak, “Farasa: A fast and furious segmenter for Arabic,” in Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations. San Diego, California: Association for Computational Linguistics, Jun. 2016, pp. 11–16. [find](https://scholar.google.com/scholar?q=A%2E+Abdelali%2C+K%2E+Darwish%2C+N%2E+Durrani%2C+and+H%2E+Mubarak%2C+%E2%80%9CFarasa%3A+A+fast+and+furious+segmenter+for+Arabic%2C%E2%80%9D+in+Proceedings+of+the+2016+Conference+of+the+North+American+Chapter+of+the+Association+for+Compu) [our card](https://md.research.elamur.ai/free/en/papers/aclbulk-2016-abdelali-farasa-fast-furious-segmenter-arabic/)
11. A. Pasha, M. Al-Badrashiny, M. Diab, A. El Kholy, R. Eskander, N. Habash, M. Pooleery, O. Rambow, and R. Roth, “MADAMIRA: A fast, comprehensive tool for morphological analysis and disambiguation of Arabic,” in Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC’14). Reykjavik, Iceland: European Language Resources Association (ELRA), May 2014, pp. 1094–1101. [find](https://scholar.google.com/scholar?q=A%2E+Pasha%2C+M%2E+Al%2DBadrashiny%2C+M%2E+Diab%2C+A%2E+El+Kholy%2C+R%2E+Eskander%2C+N%2E+Habash%2C+M%2E+Pooleery%2C+O%2E+Rambow%2C+and+R%2E+Roth%2C+%E2%80%9CMADAMIRA%3A+A+fast%2C+comprehensive+tool+for+morphological+analysis+and+disambiguation+of+Ar) [our card](https://md.research.elamur.ai/free/en/papers/lingbulk-2014-pasha-madamira-fast-comprehensive-tool-morphological/)
12. Y. Benajiba, P. Rosso, and J. M. Benedı́Ruiz, “Anersys: An arabic named entity recognition system based on maximum entropy,” in Computational Linguistics and Intelligent Text Processing, A. Gelbukh, Ed. Berlin, Heidelberg: Springer Berlin Heidelberg, 2007, pp. 143–153. [find](https://scholar.google.com/scholar?q=Y%2E+Benajiba%2C+P%2E+Rosso%2C+and+J%2E+M%2E+Bened%C4%B1%CC%81Ruiz%2C+%E2%80%9CAnersys%3A+An+arabic+named+entity+recognition+system+based+on+maximum+entropy%2C%E2%80%9D+in+Computational+Linguistics+and+Intelligent+Text+Processing%2C+A%2E+Gelbukh%2C+E)
13. T. Kudo and J. Richardson, “SentencePiece: A simple and language independent subword tokenizer and detokenizer for neural text processing,” in Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing: System Demonstrations. Brussels, Belgium: Association for Computational Linguistics, Nov. 2018, pp. 66– 71. [find](https://scholar.google.com/scholar?q=T%2E+Kudo+and+J%2E+Richardson%2C+%E2%80%9CSentencePiece%3A+A+simple+and+language+independent+subword+tokenizer+and+detokenizer+for+neural+text+processing%2C%E2%80%9D+in+Proceedings+of+the+2018+Conference+on+Empirical+Methods+i) [our card](https://md.research.elamur.ai/free/en/papers/kudo-2018-sentencepiece-simple-language-independent-subword/)
14. I. A. El-Khair, “1.5 billion words arabic corpus,” ArXiv, vol. abs/1611.04033, 2016. [↗](https://arxiv.org/abs/1611.04033)
15. I. Zeroual, D. Goldhahn, T. Eckart, and A. Lakhouaja, “OSIAN: Open source international Arabic news corpus - preparation and integration into the CLARIN-infrastructure,” in Proceedings of the Fourth Arabic Natural Language Processing Workshop. Florence, Italy: Association for Computational Linguistics, Aug. 2019, pp. 175–182. [find](https://scholar.google.com/scholar?q=I%2E+Zeroual%2C+D%2E+Goldhahn%2C+T%2E+Eckart%2C+and+A%2E+Lakhouaja%2C+%E2%80%9COSIAN%3A+Open+source+international+Arabic+news+corpus+%2D+preparation+and+integration+into+the+CLARIN%2Dinfrastructure%2C%E2%80%9D+in+Proceedings+of+the+Fourth+A)
16. T. Wolf, L. Debut, V. Sanh, J. Chaumond, C. Delangue, A. Moi, P. Cistac, T. Rault, R. Louf, M. Funtowicz, J. Davison, S. Shleifer, P. von Platen, C. Ma, Y. Jernite, J. Plu, C. Xu, T. Le Scao, S. Gugger, M. Drame, Q. Lhoest, and A. Rush, “Transformers: State-of-the-art natural language processing,” in Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations. Online: Association for Computational Linguistics, Oct. 2020, pp. 38–45. [find](https://scholar.google.com/scholar?q=T%2E+Wolf%2C+L%2E+Debut%2C+V%2E+Sanh%2C+J%2E+Chaumond%2C+C%2E+Delangue%2C+A%2E+Moi%2C+P%2E+Cistac%2C+T%2E+Rault%2C+R%2E+Louf%2C+M%2E+Funtowicz%2C+J%2E+Davison%2C+S%2E+Shleifer%2C+P%2E+von+Platen%2C+C%2E+Ma%2C+Y%2E+Jernite%2C+J%2E+Plu%2C+C%2E+Xu%2C+T%2E+Le+Scao%2C+S%2E+Gugge) [our card](https://md.research.elamur.ai/free/en/papers/wolf-2020-transformers-state-art-natural-language/)
17. M. Abdul-Mageed, A. Elmadany, and E. M. B. Nagoudi, “ARBERT & MARBERT: Deep bidirectional transformers for Arabic,” in Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). Online: Association for Computational Linguistics, Aug. 2021, pp. 7088–7105. [find](https://scholar.google.com/scholar?q=M%2E+Abdul%2DMageed%2C+A%2E+Elmadany%2C+and+E%2E+M%2E+B%2E+Nagoudi%2C+%E2%80%9CARBERT+%26+MARBERT%3A+Deep+bidirectional+transformers+for+Arabic%2C%E2%80%9D+in+Proceedings+of+the+59th+Annual+Meeting+of+the+Association+for+Computational+Lingu) [our card](https://md.research.elamur.ai/free/en/papers/abdul-2021-mageed-arbert-marbert-deep-bidirectional-transformers/)