Elamur Research⌕
☰
home / papers / 2015 / Named Entity Recognition for Arabic Social Media

Named Entity Recognition for Arabic Social Media

Paper · 2015 · Ayah Zirikly and Mona Diab

Summary

The paper addresses named entity recognition for Arabic social media, where Modern Standard Arabic (MSA) and Dialectal Arabic (DA) mix within the same message. In this setting, spelling is not standardized, capitalization does not exist, and name dictionaries quickly go out of date, especially for person nicknames. The authors build a CRF sequence labeler and start from a language-independent baseline: neighbouring words plus leading and trailing character n-grams. On top of that they add lexical, contextual and morphological features (POS, clitic and affix information, aspect, person), along with two unsupervised word-representation signals: skip-gram word2vec and Brown Clustering (BC). Gazetteer features, including fuzzy Levenshtein matching, are tested separately so the authors can check whether word representations can replace them. They evaluate on a Twitter set and on an Egyptian dialect set (DA-EGY). On DA-EGY, the best configuration without gazetteers reaches 72.68% F1, which beats the earlier gazetteer-based DA-NER system at 70.31%. This matters because high-coverage name lists for informal text are costly to build and maintain.

Key points

  • On Twitter, adding BC and word2vec raised F1 by roughly 10 points over the baseline (45.99%). BC contributed about 6 points and word2vec about 3, consistent with Turian et al.'s English finding that clusters handle rare words better.
  • Best Twitter results: 57.84% F1 without external resources (FS13) and 59.59% with gazetteers (FS12). The authors' reimplementation of Darwish and Gao's Wikipedia-gazetteer features scored 51.62%, against the 55% originally reported.
  • Levenshtein gazetteer matching (LVM-GAZ, distance thresholds 1 and 2) gave about 0.8 points over exact and partial matching (EX-GAZ, PAR-GAZ). Among the morphological features, CAPS and POS helped most, mainly for PER.
  • On DA-EGY (5-fold cross-validation, ORG dropped because it is under 0.05% of the data), FS12 with gazetteers reached 77.12% F1. Word2vec alone without BC (FS15) reached 72.61%, nearly matching the best gazetteer-free result.
  • Results are reported as precision, recall and F1, using a weighted macro-average over PER, LOC and ORG.

Interesting ideas

  • I1. Unsupervised word clusters learned from raw text can partly replace hand-built name lists, and they generalize better to rare words than dense embeddings do.
  • I2. Matching dictionary entries with a small edit-distance threshold, instead of exact strings, absorbs spelling variation when no standard orthography exists.
  • I3. A strong baseline built only from surrounding words and leading or trailing character n-grams needs no morphological analyzer, so it carries over to varieties where existing tools break.
  • I4. Scoring with precision, recall and F1 rather than accuracy avoids being fooled by the majority "not an entity" class. Dropping a class with almost no training instances also keeps the evaluation fair.
  • I5. A cumulative ablation ladder (FS1 through FS15), where each step adds one feature group, makes it clear which component produced which gain.

For Mova

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

  • I1 (tokens): Brown-style cluster IDs could serve as a cheap tagger feature, helping proper-noun and entity-type decisions on names the lexicon has never seen.
  • I2 (world): when linking entity mentions to Wikidata or WordNet entries, a bounded Levenshtein fallback could recover misspelled or variant names, with the match distance logged so the decision stays explainable.
  • I3 (tokens): affix n-gram features give the tagger a robust backoff path for noisy or informal English where the morphology module fails.
  • I4 (learn): evaluation of entity and frame extraction should report per-class P/R/F1, and should flag classes too sparse to measure, so the majority class cannot mask failures.
  • I5 (learn): new silver-data sources or repair rules could be added one at a time with scores recorded at each step, so every improvement can be traced to a specific change.

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

🎓George Washington UniversityWashington · United States51 works · 8,148 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

🧭 Topics 2Research subjects and languages the work is about.Semitic languages · 2,908 worksNamed entity recognition · 200 works

Cite

Cite as Ayah Zirikly and Mona Diab. 2015. Named Entity Recognition for Arabic Social Media. Proceedings of the 1st Workshop on Vector Space Modeling for Natural Language Processing.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.3115/v1/w15-1524

@inproceedings{zirikly2015named,
  title = {{Named Entity Recognition for Arabic Social Media}},
  author = {Ayah Zirikly and Mona Diab},
  booktitle = {Proceedings of the 1st Workshop on Vector Space Modeling for Natural Language Processing},
  year = {2015},
  doi = {10.3115/v1/w15-1524},
  url = {https://doi.org/10.3115/v1/w15-1524}
}

References

32 references, 14 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Sherief Abdallah, Khaled Shaalan, and Muhammad Shoaib. 2012. Integrating rule-based system with classification for arabic named entity recognition. In Computational Linguistics and Intelligent Text Processing, pages 311–322. Springer. find
  2. Ahmed Abdul-Hamid and Kareem Darwish. 2010. Simplified feature set for arabic named entity recognition. In Proceedings of the 2010 Named Entities Workshop, NEWS ’10, pages 110–115, Stroudsburg, PA, USA. Association for Computational Linguistics.
  3. Alfred V. Aho and Margaret J. Corasick. 1975. Efficient string matching: An aid to bibliographic search. Commun. ACM, 18(6):333–340, June. find
  4. Yassine Benajiba and Paolo Rosso. 2008. Arabic named entity recognition using conditional random fields. In Proc. of Workshop on HLT & NLP within the Arabic World, LREC, volume 8, pages 143–153. Citeseer. find
  5. Yassine Benajiba, Paolo Rosso, and José-Miguel Benedı́. 2007. Anersys: An arabic named entity recognition system based on maximum entropy. In CICLing, pages 143–153. find
  6. Yassine Benajiba, Mona Diab, and Paolo Rosso. 2008. Arabic named entity recognition using optimized feature sets. In Proceedings of the Conference on Empirical Methods in Natural Language Processing, pages 284–293. Association for Computational Linguistics.
  7. Yassine Benajiba, Mona Diab, and Paolo Rosso. 2009. Arabic named entity recognition: A feature-driven study. Audio, Speech, and Language Processing, IEEE Transactions on, 17(5):926–934. ↗
  8. Yassine Benajiba, Imed Zitouni, Mona Diab, and Paolo Rosso. 2010. Arabic named entity recognition: Using features extracted from noisy data. In Proceedings of the ACL 2010 Conference Short Papers, ACLShort ’10, pages 281–285, Stroudsburg, PA, USA. Association for Computational Linguistics.
  9. Peter F Brown, Peter V Desouza, Robert L Mercer, Vincent J Della Pietra, and Jenifer C Lai. 1992. Class-based n-gram models of natural language. Computational linguistics, 18(4):467–479. find
  10. Kareem Darwish and Wei Gao. 2014. Simple effective microblog named entity recognition: Arabic as an example. In Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC-2014), Reykjavik, Iceland, May 26-31, 2014., pages 2513–2517.
  11. Kareem Darwish, Walid Magdy, and Ahmed Mourad. 2012. Language processing for arabic microblog retrieval. In Proceedings of the 21st ACM International Conference on Information and Knowledge Management, CIKM ’12, pages 2427–2430, New York, NY, USA. ACM. find
  12. Kareem Darwish. 2013. Named entity recognition using cross-lingual resources: Arabic as an example. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1558–1567, Sofia, Bulgaria, August. Association for Computational Linguistics.
  13. Hal Daumé, III, Abhishek Kumar, and Avishek Saha. 2010. Frustratingly easy semi-supervised domain adaptation. In Proceedings of the 2010 Workshop on Domain Adaptation for Natural Language Processing, DANLP 2010, pages 53–59, Stroudsburg, PA, USA. Association for Computational Linguistics. find
  14. Ali Elsebai, Farid Meziane, and Fatma Zohra Belkredim. 2009. A rule based persons names arabic extraction system. Communications of the IBIMA, 11(6):53–59. find
  15. Sergio Ferrndez, Antonio Toral, scar Ferrndez, Antonio Ferrndez, and Rafael Muoz. 2007. Applying wikipedias multilingual knowledge to cross-lingual question answering. In In Zoubida Kedad, Nadira Lammari, Elisabeth Mtais, Farid Meziane, and Yacine Rezgui, editors, NLDB, volume 4592 of Lecture Notes in Computer Science. Springer. find
  16. Nizar Habash, Mona T Diab, and Owen Rambow. 2012. Conventional orthography for dialectal arabic. In LREC, pages 711–718.
  17. Nizar Habash, Ryan Roth, Owen Rambow, Ramy Eskander, and Nadi Tomeh. 2013. Morphological analysis and disambiguation for dialectal arabic. In Human Language Technologies: Conference of the North American Chapter of the Association of Computational Linguistics, Proceedings, June 9-14, 2013, Westin Peachtree Plaza Hotel, Atlanta, Georgia, USA, pages 426–432.
  18. Jun’ichi Kazama and Kentaro Torisawa. 2007. Exploiting wikipedia as external knowledge for named entity recognition. In Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning.
  19. John Lafferty, Andrew McCallum, and Fernando CN Pereira. 2001. Conditional random fields: Probabilistic models for segmenting and labeling sequence data. find
  20. Vladimir I Levenshtein. 1966. Binary codes capable of correcting deletions, insertions and reversals. In Soviet physics doklady, volume 10, page 707. find
  21. Tomas Mikolov, Ilya Sutskever, Kai Chen, Greg S Corrado, and Jeff Dean. 2013. Distributed representations of words and phrases and their compositionality. In Advances in Neural Information Processing Systems, pages 3111–3119. ↗
  22. David Nadeau and Satoshi Sekine. 2007. A survey of named entity recognition and classification. Lingvisticae Investigationes, 30(1):3–26. ↗
  23. David Nadeau, Peter Turney, and Stan Matwin. 2006. Unsupervised named-entity recognition: Generating gazetteers and resolving ambiguity. find
  24. Naoaki Okazaki. 2007. Crfsuite: A fast implementation of conditional random fields (crfs). find
  25. Arfath Pasha, Mohamed Al-Badrashiny, Ahmed El Kholy, Ramy Eskander, Mona Diab, Nizar Habash, Manoj Pooleery, Owen Rambow, and Ryan Roth. 2014. Madamira: A fast, comprehensive tool for morphological analysis and disambiguation of arabic. In In Proceedings of the 9th International Conference on Language Resources and Evaluation, Reykjavik, Iceland.
  26. Alan Ritter, Sam Clark, Mausam, and Oren Etzioni. 2011. Named entity recognition in tweets: An experimental study. In Proceedings of the Conference on Empirical Methods in Natural Language Processing, EMNLP ’11, pages 1524–1534, Stroudsburg, PA, USA. Association for Computational Linguistics.
  27. Khaled Shaalan and Mai Oudah. 2014. A hybrid approach to arabic named entity recognition. Journal of Information Science, 40(1):67–87. ↗
  28. Khaled Shaalan and Hafsa Raza. 2009. Nera: Named entity recognition for arabic. Journal of the American Society for Information Science and Technology, 60(8):1652–1663. ↗
  29. Khaled Shaalan. 2014. A survey of arabic named entity recognition and classification. Comput. Linguist., 40(2):469–510, June. ↗
  30. Paul Thompson and Christopher C. Dozier. 1997. Name searching and information retrieval. In In Proceedings of Second Conference on Empirical Methods in Natural Language Processing, pages 134–140. ↗
  31. Joseph Turian, Lev Ratinov, and Yoshua Bengio. 2010. Word representations: A simple and general method for semi-supervised learning. In Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics, pages 384–394. Association for Computational Linguistics.
  32. Ayah Zirikly and Mona Diab. 2014. Named entity recognition system for dialectal arabic. In Proceedings of the EMNLP 2014 Workshop on Arabic Natural Language Processing (ANLP), pages 78– 86, Doha, Qatar, October. Association for Computational Linguistics.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/zirikly-2015-named-entity-recognition-arabic-social/