--- title: "Translating Named Entities Using Monolingual and Bilingual" description: "Al-Onaizan and Knight tackle the translation of named entities (persons, locations, organizations) from… 7 references, cited by 78 papers here." authors: ["Yaser Al-Onaizan", "Kevin Knight"] year: 2002 venue: "Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics" source: "https://aclanthology.org/P02-1051/" doi: "10.3115/1073083.1073150" licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)" licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/onaizan-2002-translating-named-entities-monolingual-bilingual/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/onaizan-2002-translating-named-entities-monolingual-bilingual/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2002](https://md.research.elamur.ai/free/en/papers/2002/) / Translating Named Entities Using Monolingual and Bilingual Resources # Translating Named Entities Using Monolingual and Bilingual Resources **Authors:** [Yaser Al-Onaizan](https://md.research.elamur.ai/free/en/authors/yaser-al-onaizan/) and [Kevin Knight](https://md.research.elamur.ai/free/en/authors/kevin-knight/) · **Year:** 2002 · **Venue:** [Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2002](https://md.research.elamur.ai/free/en/venues/acl/2002/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/) **Links:** [Publisher's page](https://aclanthology.org/P02-1051/) · [PDF](https://aclanthology.org/P02-1051.pdf) · [DOI 10.3115/1073083.1073150](https://doi.org/10.3115/1073083.1073150) · [OpenAlex](https://openalex.org/W2150028966) **Citations:** 210 citations (OpenAlex, 2026-10-14) · [78](https://md.research.elamur.ai/free/en/papers/onaizan-2002-translating-named-entities-monolingual-bilingual/cited-by/) in our library, of the 141,840 papers whose bibliographies we read ## Summary Al-Onaizan and Knight tackle the translation of named entities (persons, locations, organizations) from Arabic into English. Such phrases are hard because new ones keep appearing and many are absent from bilingual dictionaries. Their system mimics a human strategy: rather than sounding out an unfamiliar name, look for English text about the same story and pick up the right spelling there. The system works in two stages. First, it generates candidates. Person names go through a transliterator that combines a phonetic model with a spelling-based letter model trained on a small name list. Location and organization names mix dictionary translations and transliterations into a regular expression, which is matched against a large English news corpus and scored with a modified IBM Model 1. Second, it re-ranks candidates using monolingual clues: Web counts, Web counts within the document's context, and co-reference with longer phrases. On a blind test set from the FBIS 2001 Multilingual Corpus, the system comes close to human translators in accuracy and clearly beats the commercial Sakhr system. It needs few costly bilingual resources. ## Key points - Overall accuracy on the blind test set: - Human: 75.30% - Sakhr (Web version): 61.30% - System top-1: 72.57% - System top-20: 84.00% - On the development set, the system's top-1 accuracy (65.20%) beat Sakhr (52.80%) but trailed the human translator (73.70%). - Person names gained the most from the system, with 77.20% top-1 accuracy on the development set versus 60.00% for the human. - Re-scoring raised overall accuracy step by step: - Development set: from 49.96% (candidate generator alone) to 65.20%. - Blind set: from 69.44% to 72.57%. - Straight Web counts gave the largest single jump. - The transliteration score linearly combines two models: - the phonetic model of Stalls and Knight (1998), which is restricted to English words with known pronunciations; - a new spelling-based model that maps letter sequences directly and adds a letter trigram model, so it can produce words never seen before. ## Interesting ideas - **Comparable text over parallel text:** use target-language documents about the same event as a source of translations. The texts do not need to be translations of each other. This fits any low-resource language pair where news is reported in several languages. - **Encode all options as one pattern:** combine every possible translation and transliteration of each word, plus optional filler words like "of" and "the," into a single regular expression. Then match it against a large monolingual corpus. The corpus acts as a cheap filter that keeps only combinations that really occur. - **Pair phonetic and spelling models:** a phonetic model alone fails on words without known pronunciations and on names that translators render letter by letter. Interpolating the two models covers both cases, a robust choice for transliterating names between any scripts. - **Generate first, then re-rank with cheap signals:** produce an n-best list, then reorder it with monolingual evidence such as Web frequency, frequency in context, and longer co-referring mentions. Since re-rankers run in sequence, their order matters, and early ones settle the easy cases. - **Report top-k alongside top-1 accuracy:** the top-k figure shows how much a better re-ranker could still gain. The authors also suggest counting only corpus hits that carry the matching NE tag, which would stop common words from inflating counts for name candidates. *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 - [University of Southern California](https://md.research.elamur.ai/free/en/universities/university-of-southern-california/) — University, Los Angeles, United States · 104 works · 33,876 citations 📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 9,086 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 8,889 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 9,296 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 9,206 works ## Cite Cite as: Yaser Al-Onaizan and Kevin Knight. 2002. Translating Named Entities Using Monolingual and Bilingual Resources. Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics. The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{alonaizan2002translating, title = {{Translating Named Entities Using Monolingual and Bilingual Resources}}, author = {Yaser Al-Onaizan and Kevin Knight}, booktitle = {Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics}, year = {2002}, doi = {10.3115/1073083.1073150}, url = {https://doi.org/10.3115/1073083.1073150} } ``` - [← previous Introduction to the CoNLL-2002 Shared Task: Language-Independent Named Entity Recognition](https://md.research.elamur.ai/free/en/papers/sang-2002-introduction-conll-2002-shared-task/) - [next → Generative Models for Statistical Parsing with Combinatory Categorial Grammar](https://md.research.elamur.ai/free/en/papers/hockenmaier-2002-generative-models-statistical-parsing-combinatory/) - [↑ back to papers of 2002](https://md.research.elamur.ai/free/en/papers/2002/) ## References 7 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Yaser Al-Onaizan and Kevin Knight. 2002. Machine Transliteration of Names in Arabic Text. In Proceedings of the ACL Workshop on Computational Approaches to Semitic Languages. [find](https://scholar.google.com/scholar?q=Yaser+Al%2DOnaizan+and+Kevin+Knight%2E+2002%2E+Machine+Transliteration+of+Names+in+Arabic+Text%2E+In+Proceedings+of+the+ACL+Workshop+on+Computational+Approaches+to+Semitic+Languages%2E) 2. Daniel M. Bikel, Richard Schwartz, and Ralph M. Weischedel. 1999. An algorithm that learns what’s in a name. Machine Learning, 34(1/3). [find](https://scholar.google.com/scholar?q=Daniel+M%2E+Bikel%2C+Richard+Schwartz%2C+and+Ralph+M%2E+Weischedel%2E+1999%2E+An+algorithm+that+learns+what%E2%80%99s+in+a+name%2E+Machine+Learning%2C+34%281%2F3%29%2E) 3. P. F. Brown, S. A. Della-Pietra, V. J. Della-Pietra, and R. L. Mercer. 1993. The Mathematics of Statistical Machine Translation: Parameter Estimation. Computational Linguistics, 19(2). [find](https://scholar.google.com/scholar?q=P%2E+F%2E+Brown%2C+S%2E+A%2E+Della%2DPietra%2C+V%2E+J%2E+Della%2DPietra%2C+and+R%2E+L%2E+Mercer%2E+1993%2E+The+Mathematics+of+Statistical+Machine+Translation%3A+Parameter+Estimation%2E+Computational+Linguistics%2C+19%282%29%2E) 4. Nancy Chinchor. 1997. MUC-7 Named Entity Task Definition. In Proceedings of the 7th Message Understanding Conference. [http://www.muc.saic.com/](http://www.muc.saic.com/). [↗](http://www.muc.saic.com/) 5. Gregory Grefenstette. 1999. The WWW as a Resource for Example-Based MT Tasks. In ASLIB’99 Translating and the Computer 21. [find](https://scholar.google.com/scholar?q=Gregory+Grefenstette%2E+1999%2E+The+WWW+as+a+Resource+for+Example%2DBased+MT+Tasks%2E+In+ASLIB%E2%80%9999+Translating+and+the+Computer+21%2E) 6. Andrei Mikheev, Marc Moens, and Calire Grover. 1999. [Named Entity Recognition without Gazetteers](https://md.research.elamur.ai/free/en/papers/mikheev-1999-named-entity-recognition-without-gazetteers/). In Proceedings of the EACL. 7. Bonnie G. Stalls and Kevin Knight. 1998. Translating Names and Technical Terms in Arabic Text. In Proceedings of the COLING/ACL Workshop on Computational Approaches to Semitic Languages. [find](https://scholar.google.com/scholar?q=Bonnie+G%2E+Stalls+and+Kevin+Knight%2E+1998%2E+Translating+Names+and+Technical+Terms+in+Arabic+Text%2E+In+Proceedings+of+the+COLING%2FACL+Workshop+on+Computational+Approaches+to+Semitic+Languages%2E) *Updated 24 October 2026*