---
title: "What in the world is a Shahab? — paper by Nezda et al. (2006)"
description: "The paper addresses the narrow coverage of most named entity recognizers. 16 references, cited by 3 papers here."
authors: ["Luke Nezda", "Andrew Hickl", "John Lehmann", "Sarmad Fayyaz"]
year: 2006
venue: "Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC'06)"
source: "https://aclanthology.org/L06-1218/"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/nezda-2006-what-world-is-shahab-wide/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/nezda-2006-what-world-is-shahab-wide/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2006](https://md.research.elamur.ai/free/en/papers/2006/page/2/) / What in the world is a Shahab?: Wide Coverage Named Entity Recognition for Arabic

# What in the world is a Shahab?: Wide Coverage Named Entity Recognition for Arabic

**Authors:** Luke Nezda, [Andrew Hickl](https://md.research.elamur.ai/free/en/authors/andrew-hickl/), John Lehmann and Sarmad Fayyaz · **Year:** 2006 · **Venue:** [Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC'06)](https://md.research.elamur.ai/free/en/venues/lrec/) ([LREC 2006](https://md.research.elamur.ai/free/en/venues/lrec/2006/)) · **Publisher:** [European Language Resources Association (ELRA)](https://md.research.elamur.ai/free/en/publishers/european-language-resources-association-elra/) · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/L06-1218/) · [PDF](https://aclanthology.org/L06-1218.pdf) · [OpenAlex](https://openalex.org/W2756800091)

**Citations:** 10 citations (OpenAlex, 2026-10-14) · [3](https://md.research.elamur.ai/free/en/papers/nezda-2006-what-world-is-shahab-wide/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
The paper addresses the narrow coverage of most named entity recognizers. MUC-era systems handled up to 8 types and CoNLL systems only 4. Prior Arabic NER work was also very limited: TAGARAB covered 5 types on a small corpus. The authors built CiceroArabic, a supervised NER system for Modern Standard Arabic that distinguishes 18 entity classes. They first designed a language-independent annotation schema drawing on MUC, DARPA TIDES and ACE guidelines. They then ran a four-month annotation project, with quality-control measures to keep annotators consistent, producing the CiceroArabic NER Corpus of about 800,000 words of newswire. The classifier uses only learned features and needs no hand-written grammars, morphological information or specialized lexicons. Trained on 600,000 words, it averages above 85% F across the 18 types. Accuracy on the core types stays about the same whether the label set is small or large. The work shows that a fine-grained label inventory need not cost accuracy if enough consistent training data exists.

## Key points
- The CiceroArabic NER Corpus contains roughly 800,000 annotated words. By comparison, the ACE 2005 Arabic release has about 133,000 words with 7 types and 45 subtypes.
- The best-scoring classes were dates (97.87% F), time quantities (90.17%), numbers (89.98%) and political locations (89.27%).
- The authors grew the training data from 100K to 300K to 600K words. At 600K, the 18-class scheme reached 85.51 F, MUC-7 (8 types) reached 85.54 and a modified CoNLL scheme (3 types) reached 82.79. The 18-class scheme showed no significant drop.
- Across 8 groupings of mention types (NAM, NUM, TIM, MISC), average F on the core types PER, ORG, GPE and LOC varied by only 0.37%.
- In the confusion analysis, most misclassifications stayed inside the correct mention-type family: 68% for NAM, 53% for NUM and 79% for TIM. MISC was the exception, with only 1.6% of errors staying inside the family.

## Interesting ideas
- The authors judge an entity inventory by three tests: annotators can tell the classes apart without domain knowledge or interpretation, the classes barely overlap, and the full set is small enough to label at once.
- The fine-grained labels are mapped onto standard coarser schemes, for example political plus geographic location becoming MUC LOCATION, so one system can be scored against older benchmarks.
- Errors are grouped by super-category to see whether the system confuses near neighbours or makes cross-family mistakes, which separates fine-distinction problems from deeper failures.
- Learning curves are plotted at several data sizes for each label granularity, checking whether adding classes costs accuracy at every data size.
- An ablation over groups of surface features (word, stem, prefix, suffix) shows how much each layer of form information contributes.

## For Mova

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

- **I1** (world): We could apply these three tests when deciding which entity kinds the text world tracks, and drop or merge ontology types that annotators or witness parsers cannot reliably separate.
- **I2** (learn): Writing explicit label-collapse mappings would let us score Mova's richer entity and role labels against standard coarse benchmarks without retraining.
- **I3** (learn): Evaluation reports could split frame or entity errors into within-family and cross-family errors, which would point repair rules at the right level.
- **I4** (learn): Before refining a tag set in silver data, we could compare curves at a few corpus sizes to confirm that finer labels do not hurt the core categories.
- **I5** (tokens): Running group-wise ablations on the tagger's word, lemma and affix features would show which morphology signals actually earn their place.

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

## Institutions

- [Language Computer](https://md.research.elamur.ai/free/en/companies/language-computer/) — Company, Richardson, United States · 9 works · 1,157 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [🧬 Biochemistry, Genetics and Molecular Biology](https://md.research.elamur.ai/free/en/shelves/biochemistry-genetics-and-molecular-biology/) · 234 works · [🧬 Molecular Biology](https://md.research.elamur.ai/free/en/shelves/biochemistry-genetics-and-molecular-biology/molecular-biology/) · 231 works

🧭 Topics (2): [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/) · 2,908 works · [🏷️ Named entity recognition](https://md.research.elamur.ai/free/en/topics/named-entity-recognition/) · 200 works

## Cite

Cite as: Luke Nezda, Andrew Hickl, John Lehmann and Sarmad Fayyaz. 2006. What in the world is a Shahab?: Wide Coverage Named Entity Recognition for Arabic. Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC'06).

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

```bibtex
@inproceedings{nezda2006what,
  title = {{What in the world is a Shahab?: Wide Coverage Named Entity Recognition for Arabic}},
  author = {Luke Nezda, Andrew Hickl, John Lehmann and Sarmad Fayyaz},
  booktitle = {Proceedings of the Fifth International Conference on Language Resources and Evaluation (LREC'06)},
  year = {2006},
  doi = {10.63317/4wemjmnwuzzu},
  url = {https://doi.org/10.63317/4wemjmnwuzzu}
}
```

- [← previous Lexicon Development for Varieties of Spoken Colloquial Arabic](https://md.research.elamur.ai/free/en/papers/graff-2006-lexicon-development-varieties-spoken-colloquial/)
- [next → Arabic WordNet and the Challenges of Arabic](https://md.research.elamur.ai/free/en/papers/elkateb-2006-arabic-wordnet-challenges-arabic/)
- [↑ back to papers of 2006](https://md.research.elamur.ai/free/en/papers/2006/page/2/)

## References

16 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Douglas Appelt, Jerry Hobbs, John Bear, David Israel, Megumi Kameyama, Andy Kehler, David Martin, Karen Meyers, and Mabry Tyson. 1995. Sri international fastus system: Muc-6 test results and analysis. In In Proceedings of the Sixth Message Understanding Conf. (MUC-6). Morgan Kaufman. [↗](https://doi.org/10.21236/ada460970)
2. D. Bikel, S. Miller, R. Schwartz, and R. Weischedel. 1997. [Nymble: A high-performance learning name-finder](https://md.research.elamur.ai/free/en/papers/bikel-1997-nymble-high-performance-learning-name/). In Proceedings of the Fifth Conference on Applied Natural Language Processing. [↗](https://doi.org/10.48550/arxiv.cmp-lg/9803003)
3. Nancy Chinchor, 1997. MUC-7 Named Entity Task Definition. [find](https://scholar.google.com/scholar?q=Nancy+Chinchor%2C+1997%2E+MUC%2D7+Named+Entity+Task+Definition%2E)
4. S. Cucerzan and D. Yarowsky. 1999. Language independent named entity recognition combining morphological and contextual evidence. In In Proceedings of the Joint SIGDAT Conference on Empirical Methods in Natural Language Processing and Very Large Corpora (EMNLP/VLC-99). [find](https://scholar.google.com/scholar?q=S%2E+Cucerzan+and+D%2E+Yarowsky%2E+1999%2E+Language+independent+named+entity+recognition+combining+morphological+and+contextual+evidence%2E+In+In+Proceedings+of+the+Joint+SIGDAT+Conference+on+Empirical+Methods+)
5. Mona Diab, Kadri Hacioglu, and Daniel Jurafsky. 2004. [Automatic tagging of arabic text: From raw text to base phrase chunks](https://md.research.elamur.ai/free/en/papers/diab-2004-automatic-tagging-arabic-text-raw/). In Proceedings of HLT-NAACL 2004.
6. Jan Hajic, Otakar Smrz, Petr Zemanek, Petr Pajas, Jan Snaidauf, Emanuel Beska, Jakub Kracmar, and Kamila Hassanova. 2004. Prague Arabic Dependency Bank 1.0. Number 1.0 LDC2004T23. [find](https://scholar.google.com/scholar?q=Jan+Hajic%2C+Otakar+Smrz%2C+Petr+Zemanek%2C+Petr+Pajas%2C+Jan+Snaidauf%2C+Emanuel+Beska%2C+Jakub+Kracmar%2C+and+Kamila+Hassanova%2E+2004%2E+Prague+Arabic+Dependency+Bank+1%2E0%2E+Number+1%2E0+LDC2004T23%2E)
7. Sanda Harabagiu, Dan Moldovan, Marius Pasca, Rada Mihalcea, Mihai Surdeanu, Razvan Bunsecu, Roxana Girju, Vasile Rus, and Paul Morarescu. 2001. The role of lexico-semantic feedback in open-domain textual question-answering. In ACL01. [find](https://scholar.google.com/scholar?q=Sanda+Harabagiu%2C+Dan+Moldovan%2C+Marius+Pasca%2C+Rada+Mihalcea%2C+Mihai+Surdeanu%2C+Razvan+Bunsecu%2C+Roxana+Girju%2C+Vasile+Rus%2C+and+Paul+Morarescu%2E+2001%2E+The+role+of+lexico%2Dsemantic+feedback+in+open%2Ddomain+text)
8. Dan Klein, Joseph Smarr, Huy Nguyen, and Chris Manning. 2003. Named entity recognition with character-level models. In Conference on Natural Language Learning ’03. [find](https://scholar.google.com/scholar?q=Dan+Klein%2C+Joseph+Smarr%2C+Huy+Nguyen%2C+and+Chris+Manning%2E+2003%2E+Named+entity+recognition+with+character%2Dlevel+models%2E+In+Conference+on+Natural+Language+Learning+%E2%80%9903%2E)
9. L. Larkey, L. Ballesteros, and M. Connell. 2002. Improving stemming for arabic information retrieval: Light stemming and co-occurrence analysis. In SIGIR. [find](https://scholar.google.com/scholar?q=L%2E+Larkey%2C+L%2E+Ballesteros%2C+and+M%2E+Connell%2E+2002%2E+Improving+stemming+for+arabic+information+retrieval%3A+Light+stemming+and+co%2Doccurrence+analysis%2E+In+SIGIR%2E)
10. Linguistic Data Consortium, 2005. ACE (Automatic Content Extraction) Arabic Annotation Guidelines for Entities. [find](https://scholar.google.com/scholar?q=Linguistic+Data+Consortium%2C+2005%2E+ACE+%28Automatic+Content+Extraction%29+Arabic+Annotation+Guidelines+for+Entities%2E)
11. Mohamed Maamouri, Ann Bies, Tim Buckwalter, and Wigdan Mekki. 2004. The penn arabic treebank: Building a large-scale annotated arabic corpus. In NEMLAR International Conference on Arabic Language Resources and Tools. [find](https://scholar.google.com/scholar?q=Mohamed+Maamouri%2C+Ann+Bies%2C+Tim+Buckwalter%2C+and+Wigdan+Mekki%2E+2004%2E+The+penn+arabic+treebank%3A+Building+a+large%2Dscale+annotated+arabic+corpus%2E+In+NEMLAR+International+Conference+on+Arabic+Language+Reso)
12. John Maloney and Michael Niv. 1998. Tagarab: A fast, accurate arabic name recogniser using high precision morphological analysis. In Proceedings of the Workshop on Computational Approaches to Semitic Languages. [find](https://scholar.google.com/scholar?q=John+Maloney+and+Michael+Niv%2E+1998%2E+Tagarab%3A+A+fast%2C+accurate+arabic+name+recogniser+using+high+precision+morphological+analysis%2E+In+Proceedings+of+the+Workshop+on+Computational+Approaches+to+Semitic+)
13. Marius Pasca. 2004. Acquisition of categorized named entities for web search. In Proceedings of the Thirteenth ACM conference on Information and knowledge management. [find](https://scholar.google.com/scholar?q=Marius+Pasca%2E+2004%2E+Acquisition+of+categorized+named+entities+for+web+search%2E+In+Proceedings+of+the+Thirteenth+ACM+conference+on+Information+and+knowledge+management%2E)
14. Mihai Surdeanu and Sanda Harabagiu. 2002. Infrastructure for open-domain information extraction. In Proceedings of the Human Language Technology Conference. [find](https://scholar.google.com/scholar?q=Mihai+Surdeanu+and+Sanda+Harabagiu%2E+2002%2E+Infrastructure+for+open%2Ddomain+information+extraction%2E+In+Proceedings+of+the+Human+Language+Technology+Conference%2E)
15. M. Thelen and E. Riloff. 2002. [A bootstrapping method for learning semantic lexicons using extraction pattern contexts](https://md.research.elamur.ai/free/en/papers/thelen-2002-bootstrapping-method-learning-semantic-lexicons/). In Proceedings of the 2002 Conference on Empirical Methods in Natural Language Processing.
16. Erik F. Tjong Kim Sang and Fien De Meulder. 2003. [Introduction to the conll-2003 shared task: Language-independent named entity recognition](https://md.research.elamur.ai/free/en/papers/sang-2003-introduction-conll-2003-shared-task/). In Walter Daelemans and Miles Osborne, editors, Proceedings of CoNLL-2003. [↗](https://doi.org/10.48550/arxiv.cs/0306050)

*Updated 1 November 2026*