---
title: "AMWAL: Named Entity Recognition for Arabic Financial News"
description: "Arabic NER tools mostly cover generic types such as people, organizations and places. 26 references."
authors: ["Muhammad S. Abdo", "Yash A. Hatekar", "Damir Cavar"]
year: 2025
venue: "Proceedings of the Joint Workshop of the 9th Financial Technology and Natural Language Processing (FinNLP), the 6th Financial Narrative Processing (FNP), and the 1st Workshop on Large Language Models for Finance and Legal (LLMFinLegal)"
source: "https://aclanthology.org/2025.finnlp-1.20/"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/abdo-2025-amwal-named-entity-recognition-arabic/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/abdo-2025-amwal-named-entity-recognition-arabic/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2025](https://md.research.elamur.ai/free/en/papers/2025/page/16/) / AMWAL: Named Entity Recognition for Arabic Financial News

# AMWAL: Named Entity Recognition for Arabic Financial News

**Authors:** [Muhammad S. Abdo](https://md.research.elamur.ai/free/en/authors/muhammad-s-abdo/), [Yash A. Hatekar](https://md.research.elamur.ai/free/en/authors/yash-a-hatekar/) and Damir Cavar · **Year:** 2025 · **Venue:** Proceedings of the Joint Workshop of the 9th Financial Technology and Natural Language Processing (FinNLP), the 6th Financial Narrative Processing (FNP), and the 1st Workshop on Large Language Models for Finance and Legal (LLMFinLegal) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2025.finnlp-1.20/) · [PDF](https://aclanthology.org/2025.finnlp-1.20.pdf)

## Summary
Arabic NER tools mostly cover generic types such as people, organizations and places. Outside medicine, almost nothing targets a specific domain. AMWAL fills this gap for financial news. The authors gathered 26,231 articles (about 9.8 million tokens, 2000–2023) from three Arabic financial outlets and normalized the orthography by removing diacritics and unifying alef, yaa and taa-marbuta variants. They took the label set mainly from FIBO and added six types of their own, for 20 categories. Annotation was semi-automatic. Hearst-style lexical queries in Textometry pulled out entities already typed by their hypernym, frequency-based keyword searches expanded these, and a manual review followed. This yielded 17,185 entities. A spaCy NER pipeline on top of AraBERT Large reached macro P/R/F1 of 96.08/95.87/95.97 on a held-out 20% of files. That is well above CAMeL (91) and Wojood (80), two general models that cannot see most finance-specific types. The work gives Arabic its first financial NER resource and released model and training files.

## Key points
- Corpus sources: Almal News (11,012 articles), Al-Sharq (8,106) and the Aljazeera business section (2,627). The text is MSA only, and the authors say the model may not transfer to dialectal blogs or social media.
- The 20 labels come from FIBO plus six added types: GEOPOLITICAL, BANK, METRIC, STOCK EXCHANGE, MEDIA, FINANCIAL MARKET. CORPORATION dominates with 6,840 of 17,185 entities, while MEDIA is the rarest with 103.
- Training setup:
  - spaCy pipeline with AraBERT Large
  - batch size 50, dropout 0.1
  - up to 20,000 steps, early-stopping patience 1600
  - 80/20 split at the file level (20,984 / 5,247 JSON files)
- Per-type F1 is near 98–99 for CURRENCY, TIME and EVENT. It is weakest for PERSON (80), CORP (81), CITY (81) and ORG (85). The baselines score 0 on several finance-specific labels, which drags down their macro averages.
- Errors come from names that contain words of another category, such as a nationality inside a company or person name. Homographs also cause errors: "Nissan" is spelled like the Levantine word for April. The authors admit that much of the score may come from recognizing entities already seen in training.

## Interesting ideas
- Patterns like "banks such as X" or "X and Y bank" harvest entity strings and their type in one step, because the hypernym in the pattern supplies the label; a minimum-frequency cutoff (5 occurrences) filters noise.
- After the first harvest, the most frequent head tokens and bigrams among found entities become new search keys, catching the short name, full legal name and "X company" forms of one entity.
- Rather than inventing a label set ad hoc, the authors anchor categories in an industry-standard ontology (FIBO) and only add types the domain clearly needs.
- The authors warn that high scores may reflect memorized names, since the split separates files but not entity strings.
- Many errors arise when parts of a multiword name carry their own type, such as a nationality inside a company name or a month-like brand.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (world): run "such as / and other" patterns over parsed English text to collect typed instances, then check them against WordNet/Wikidata to grow the ontology with corpus-attested members.
- **I2** (world): use the same frequency-driven expansion to build alias sets per entity, so the world of a text merges "Orascom" and "Orascom Construction PLC"-style variants into one holder of states.
- **I3** (world): when adding domain entity types, map them onto an existing external ontology first, so every type stays explainable and linkable instead of being a private tag.
- **I4** (learn): add an evaluation split where test entity strings never appear in training, as a negative control against memorization in any silver-trained tagger.
- **I5** (tokens): recognize multiword proper names as one unit before semantic typing, so inner words like "Swedish" or "Italy" are not typed on their own.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

📚 Shelves (3): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [📈 Economics, Econometrics and Finance](https://md.research.elamur.ai/free/en/shelves/economics-econometrics-and-finance/) · 23 works

🧭 Topics (2): [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/) · 3,066 works · [🏷️ Named entity recognition](https://md.research.elamur.ai/free/en/topics/named-entity-recognition/) · 214 works

## Cite

Cite as: Muhammad S. Abdo, Yash A. Hatekar and Damir Cavar. 2025. AMWAL: Named Entity Recognition for Arabic Financial News. Proceedings of the Joint Workshop of the 9th Financial Technology and Natural Language Processing (FinNLP), the 6th Financial Narrative Processing (FNP), and the 1st Workshop on Large Language Models for Finance and Legal (LLMFinLegal).

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

```bibtex
@inproceedings{abdo2025amwal,
  title = {{AMWAL: Named Entity Recognition for Arabic Financial News}},
  author = {Muhammad S. Abdo, Yash A. Hatekar and Damir Cavar},
  booktitle = {Proceedings of the Joint Workshop of the 9th Financial Technology and Natural Language Processing (FinNLP), the 6th Financial Narrative Processing (FNP), and the 1st Workshop on Large Language Models for Finance and Legal (LLMFinLegal)},
  year = {2025},
  url = {https://aclanthology.org/2025.finnlp-1.20/}
}
```

- [← previous Amharic News Topic Classification: Dataset and Transformer-Based Model Benchmarks](https://md.research.elamur.ai/free/en/papers/marilign-2025-amharic-news-topic-classification-dataset/)
- [next → An Empirical Study of Chinese Entity Linking Based on Large Language Model](https://md.research.elamur.ai/free/en/papers/xu-2025-empirical-study-chinese/)
- [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/16/)

## References

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

1. Merve Alanyali, Helen Susannah Moat, and Tobias Preis. 2013. Quantifying the relationship between financial news and the stock market. Scientific reports, 3(1):3578. [find](https://scholar.google.com/scholar?q=Merve+Alanyali%2C+Helen+Susannah+Moat%2C+and+Tobias+Preis%2E+2013%2E+Quantifying+the+relationship+between+financial+news+and+the+stock+market%2E+Scientific+reports%2C+3%281%29%3A3578%2E)
2. Wissam Antoun, Fady Baly, and Hazem Hajj. 2020. [Arabert: Transformer-based model for arabic language understanding](https://md.research.elamur.ai/free/en/papers/antoun-2020-arabert-transformer-based-model-arabic/). arXiv preprint arXiv:2003.00104. [↗](https://arxiv.org/abs/2003.00104)
3. Mike Bennett. 2013. The financial industry business ontology: Best practice for big data. Journal of Banking Regulation, 14(3):255–268. [find](https://scholar.google.com/scholar?q=Mike+Bennett%2E+2013%2E+The+financial+industry+business+ontology%3A+Best+practice+for+big+data%2E+Journal+of+Banking+Regulation%2C+14%283%29%3A255%E2%80%93268%2E)
4. Serge Heiden. 2010. The txm platform: Building open-source textual analysis software compatible with the tei encoding scheme. In 24th Pacific Asia conference on language, information and computation, volume 2, pages 389–398. Institute for Digital Enhancement of Cognitive Development, Waseda University. [find](https://scholar.google.com/scholar?q=Serge+Heiden%2E+2010%2E+The+txm+platform%3A+Building+open%2Dsource+textual+analysis+software+compatible+with+the+tei+encoding+scheme%2E+In+24th+Pacific+Asia+conference+on+language%2C+information+and+computation%2C+)
5. Lars Hillebrand, Tobias Deußer, Tim Dilmaghani, Bernd Kliem, Rüdiger Loitz, Christian Bauckhage, and Rafet Sifa. 2022. Kpi-bert: A joint named entity recognition and relation extraction model for financial reports. In 2022 26th International Conference on Pattern Recognition (ICPR), pages 606–612. IEEE. [find](https://scholar.google.com/scholar?q=Lars+Hillebrand%2C+Tobias+Deu%C3%9Fer%2C+Tim+Dilmaghani%2C+Bernd+Kliem%2C+R%C3%BCdiger+Loitz%2C+Christian+Bauckhage%2C+and+Rafet+Sifa%2E+2022%2E+Kpi%2Dbert%3A+A+joint+named+entity+recognition+and+relation+extraction+model+for+fina)
6. Matthew Honnibal and Ines Montani. 2017. spacy 2: Natural language understanding with bloom embeddings, convolutional neural networks and incremental parsing. To appear, 7(1):411–420. [find](https://scholar.google.com/scholar?q=Matthew+Honnibal+and+Ines+Montani%2E+2017%2E+spacy+2%3A+Natural+language+understanding+with+bloom+embeddings%2C+convolutional+neural+networks+and+incremental+parsing%2E+To+appear%2C+7%281%29%3A411%E2%80%93420%2E)
7. Nada Boudjellal, Huaping Zhang, Asif Khan, Arshad Ahmad, Rashid Naseem, Jianyun Shang, and Lin Dai. 2021. Abioner: a bert-based model for arabic biomedical named-entity recognition. Complexity, 2021:1–6. [find](https://scholar.google.com/scholar?q=Nada+Boudjellal%2C+Huaping+Zhang%2C+Asif+Khan%2C+Arshad+Ahmad%2C+Rashid+Naseem%2C+Jianyun+Shang%2C+and+Lin+Dai%2E+2021%2E+Abioner%3A+a+bert%2Dbased+model+for+arabic+biomedical+named%2Dentity+recognition%2E+Complexity%2C+2021%3A1)
8. Ann Devitt and Khurshid Ahmad. 2007. Sentiment polarity identification in financial news: A cohesion-based approach. In Proceedings of the 45th annual meeting of the association of computational linguistics, pages 984–991. [find](https://scholar.google.com/scholar?q=Ann+Devitt+and+Khurshid+Ahmad%2E+2007%2E+Sentiment+polarity+identification+in+financial+news%3A+A+cohesion%2Dbased+approach%2E+In+Proceedings+of+the+45th+annual+meeting+of+the+association+of+computational+lingu)
9. Duygu Dinç. 2022. Financial named entity recognition for turkish news texts. Master’s thesis, Middle East Technical University. [find](https://scholar.google.com/scholar?q=Duygu+Din%C3%A7%2E+2022%2E+Financial+named+entity+recognition+for+turkish+news+texts%2E+Master%E2%80%99s+thesis%2C+Middle+East+Technical+University%2E)
10. EDM Council and Object Management Group, Inc. 2017. Financial Industry Business Ontology – Indices and Indicators. OMG Document Number: formal/2017-07-01, Release Date: July 2017, Normative Reference: [http://www.omg.org/spec/EDMC-FIBO/IND/1.0/](http://www.omg.org/spec/EDMC-FIBO/IND/1.0/). [↗](http://www.omg.org/spec/EDMC-FIBO/IND/1.0/)
11. Tome Eftimov, Barbara Koroušić Seljak, and Peter Korošec. 2017. A rule-based named-entity recognition method for knowledge extraction of evidence-based dietary recommendations. PloS one, 12(6):e0179488. [find](https://scholar.google.com/scholar?q=Tome+Eftimov%2C+Barbara+Korou%C5%A1ic%CC%81+Seljak%2C+and+Peter+Koro%C5%A1ec%2E+2017%2E+A+rule%2Dbased+named%2Dentity+recognition+method+for+knowledge+extraction+of+evidence%2Dbased+dietary+recommendations%2E+PloS+one%2C+12%286%29%3Ae01794)
12. Dimitra Farmakiotou, Vangelis Karkaletsis, John Kout-sias, George Sigletos, Constantine D Spyropoulos, and Panagiotis Stamatopoulos. 2000. Rule-based named entity recognition for greek financial texts. In Proceedings of the Workshop on Computational lexicography and Multimedia Dictionaries (COMLEX 2000), pages 75–78. [find](https://scholar.google.com/scholar?q=Dimitra+Farmakiotou%2C+Vangelis+Karkaletsis%2C+John+Kout%2Dsias%2C+George+Sigletos%2C+Constantine+D+Spyropoulos%2C+and+Panagiotis+Stamatopoulos%2E+2000%2E+Rule%2Dbased+named+entity+recognition+for+greek+financial+texts)
13. Rema Muftah Hamad and Ahmed Mohamed Abushaala. 2023. Medical named entity recognition in arabic text using svm. In 2023 IEEE 3rd International Maghreb Meeting of the Conference on Sciences and Techniques of Automatic Control and Computer Engineering (MI-STA), pages 200–205. IEEE. [find](https://scholar.google.com/scholar?q=Rema+Muftah+Hamad+and+Ahmed+Mohamed+Abushaala%2E+2023%2E+Medical+named+entity+recognition+in+arabic+text+using+svm%2E+In+2023+IEEE+3rd+International+Maghreb+Meeting+of+the+Conference+on+Sciences+and+Techniq)
14. Yash Hatekar and Muhammad Abdo. 2023. [Iunadi at nadi 2023 shared task: Country-level arabic dialect classification in tweets for the shared task nadi 2023](https://md.research.elamur.ai/free/en/papers/hatekar-2023-iunadi-nadi-2023-shared-task/). In Proceedings of ArabicNLP 2023, pages 665–669.
15. Go Inoue, Bashar Alhafni, Nurpeiis Baimukan, Houda Bouamor, and Nizar Habash. 2021. [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/). In Proceedings of the Sixth Arabic Natural Language Processing Workshop, Kyiv, Ukraine (Online). Association for Computational Linguistics.
16. Ali Jabbari, Olivier Sauvage, Hamada Zeine, and Hamza Chergui. 2020. A french corpus and annotation schema for named entity recognition and relation extraction of financial news. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 2293–2299. [find](https://scholar.google.com/scholar?q=Ali+Jabbari%2C+Olivier+Sauvage%2C+Hamada+Zeine%2C+and+Hamza+Chergui%2E+2020%2E+A+french+corpus+and+annotation+schema+for+named+entity+recognition+and+relation+extraction+of+financial+news%2E+In+Proceedings+of+the)
17. Mustafa Jarrar, Muhammad Abdul-Mageed, Mohammed Khalilia, Bashar Talafha, AbdelRahim Elmadany, Nagham Hamad, and Alaa’ Omar. 2023. [Wojoodner 2023: The first arabic named entity recognition shared task](https://md.research.elamur.ai/free/en/papers/jarrar-2023-wojoodner-2023-first-arabic-named/). arXiv preprint arXiv:2310.16153. [↗](https://arxiv.org/abs/2310.16153)
18. Mustafa Jarrar, Mohammed Khalilia, and Sana Ghanem. 2022. [Wojood: Nested arabic named entity corpus and recognition using bert](https://md.research.elamur.ai/free/en/papers/jarrar-2022-wojood-nested-arabic-named-entity/). arXiv preprint arXiv:2205.09651. [↗](https://arxiv.org/abs/2205.09651)
19. Sunisth Kumar, Davide Liu, and Alexandre Boulenger. 2023. Cross-lingual ner for financial transaction data in low-resource languages. arXiv preprint arXiv:2307.08714. [↗](https://arxiv.org/abs/2307.08714)
20. Hamada Nayel, Nourhan Marzouk, and Ahmed Elsawy. 2023. Named entity recognition for arabic medical texts using deep learning models. In 2023 Intelligent Methods, Systems, and Applications (IMSA), pages 281–285. IEEE. [find](https://scholar.google.com/scholar?q=Hamada+Nayel%2C+Nourhan+Marzouk%2C+and+Ahmed+Elsawy%2E+2023%2E+Named+entity+recognition+for+arabic+medical+texts+using+deep+learning+models%2E+In+2023+Intelligent+Methods%2C+Systems%2C+and+Applications+%28IMSA%29%2C+page)
21. Rebecca J Passonneau, Tifara Ramelson, and Boyi Xie. 2015. Named entity recognition from financial press releases. In Knowledge Discovery, Knowledge Engineering and Knowledge Management: 6th International Joint Conference, IC3K 2014, Rome, Italy, October 21-24, 2014, Revised Selected Papers 6, pages 240–254. Springer. [find](https://scholar.google.com/scholar?q=Rebecca+J+Passonneau%2C+Tifara+Ramelson%2C+and+Boyi+Xie%2E+2015%2E+Named+entity+recognition+from+financial+press+releases%2E+In+Knowledge+Discovery%2C+Knowledge+Engineering+and+Knowledge+Management%3A+6th+Internati)
22. GG Petrova, AF Tuzovsky, and Nataliya Valerievna Aksenova. 2017. Application of the financial industry business ontology (fibo) for development of a financial organization ontology. In Journal of Physics: Conference Series, volume 803, page 012116. IOP Publishing. [find](https://scholar.google.com/scholar?q=GG+Petrova%2C+AF+Tuzovsky%2C+and+Nataliya+Valerievna+Aksenova%2E+2017%2E+Application+of+the+financial+industry+business+ontology+%28fibo%29+for+development+of+a+financial+organization+ontology%2E+In+Journal+of+Phys)
23. Xiaoye Qu, Yingjie Gu, Qingrong Xia, Zechang Li, Zhefeng Wang, and Baoxing Huai. 2023. [A survey on arabic named entity recognition: Past, recent advances, and future trends](https://md.research.elamur.ai/free/en/papers/qu-2023-survey-arabic-named-entity-recognition/). IEEE Transactions on Knowledge and Data Engineering.
24. Robert P Schumaker and Hsinchun Chen. 2009. A quantitative stock prediction system based on financial news. Information Processing & Management, 45(5):571–583. [find](https://scholar.google.com/scholar?q=Robert+P+Schumaker+and+Hsinchun+Chen%2E+2009%2E+A+quantitative+stock+prediction+system+based+on+financial+news%2E+Information+Processing+%26+Management%2C+45%285%29%3A571%E2%80%93583%2E)
25. Pir Dino Soomro, Sanotsh Kumar, Arsalan Ali Shaikh, Hans Raj, et al. 2017. Bio-ner: biomedical named entity recognition using rule-based and statistical learners. International Journal of Advanced Computer Science and Applications, 8(12). [find](https://scholar.google.com/scholar?q=Pir+Dino+Soomro%2C+Sanotsh+Kumar%2C+Arsalan+Ali+Shaikh%2C+Hans+Raj%2C+et+al%2E+2017%2E+Bio%2Dner%3A+biomedical+named+entity+recognition+using+rule%2Dbased+and+statistical+learners%2E+International+Journal+of+Advanced+Com)
26. Shuwei Wang, Ruifeng Xu, Bin Liu, Lin Gui, and Yu Zhou. 2014. Financial named entity recognition based on conditional random fields and information entropy. In 2014 international conference on machine learning and cybernetics, volume 2, pages 838– 843. IEEE. [find](https://scholar.google.com/scholar?q=Shuwei+Wang%2C+Ruifeng+Xu%2C+Bin+Liu%2C+Lin+Gui%2C+and+Yu+Zhou%2E+2014%2E+Financial+named+entity+recognition+based+on+conditional+random+fields+and+information+entropy%2E+In+2014+international+conference+on+machine)

*Updated 2 November 2026*