Elamur Research⌕
☰
home / papers / 2019 / The Extended Arabic WordNet: a Case Study and an Evaluation using a Word Sense Disambiguation System

The Extended Arabic WordNet: a Case Study and an Evaluation using a Word Sense Disambiguation System

Paper · 2019 · Mohamed Ali Batita and Mounir Zrigui

Summary

Arabic WordNet (AWN) is a central lexical resource for Arabic, but it has been updated only rarely. Its 2015 extended version (Ex.V) added many words, yet it contains far fewer semantic relations than earlier releases. The authors compare the three AWN releases with Princeton WordNet. They also recap their own earlier work, which added antonym relations using corpus patterns filtered by LogDice in Sketch Engine, and derivational relations using expert-written rules based on POS switches. They discuss how such new relations can be validated and conclude that human evaluation is still the most dependable option. For an extrinsic test, they plug each AWN version into a simple lexical-chain WSD algorithm adapted from Galley et al. and run it on the Khaleej-2004 corpus. Only the resource changes between runs. With the newly added relations, precision reaches 78.6% and F1 74.6%, clearly above the other versions. The paper argues that relation density matters as much as vocabulary coverage, and that downstream tasks are a practical way to judge a lexical resource.

Key points

  • AWN releases:
    • V1 (2006): 9,698 synsets, 21,813 words, 6 relation types.
    • V2 (2010): 11,269 synsets, 23,481 words, 22 relation types, 161,705 relations.
    • Ex.V (Regragui et al., LMF format, available via Open Multilingual Wordnet): 8,550 synsets and 60,157 words (including 2,650 broken plurals and 14,683 verbal roots), but only about 41K relations of 5 types.
  • Ex.V originally contained just 14 antonym pairs. The authors' derivational work produced 8 new relation types, such as a verb-to-agent-noun link (kataba → kaātibun), all checked by a lexicographer.
  • The WSD algorithm builds a graph whose nodes are senses and whose weighted edges come from relations between senses. Weights depend on relation type and on the distance between words, and the highest-scoring sense wins. The original algorithm uses 4 relation types; the authors extended it to accept every relation in Ex.V.
  • Evaluation corpus: Khaleej-2004, with 5,690 news documents in 4 categories and roughly 3M words.
  • Precision results:
    • V2: 69.2%.
    • Ex.V without the new relations: 72.7% (recall 66.9%, F1 69.6%).
    • Ex.V with the new relations: 78.6% (recall 71.1%, F1 74.6%).

Interesting ideas

  • I1. Judge a lexical resource extrinsically: keep the downstream algorithm deliberately simple and unchanged, swap only the resource, and treat any score difference as the resource's contribution.
  • I2. Count relations as well as entries: a release with more words but sparse links can perform worse than a smaller, densely linked one.
  • I3. Derive relations from morphology using expert rules keyed on POS switches (verb → agent noun, verb → instrument), so lexical links follow systematic word formation.
  • I4. Harvest antonym candidates with lexical patterns over a corpus, rank them with an association score like LogDice, and send only top-ranked pairs to manual review.
  • I5. Disambiguate senses with a sense graph, where each candidate's score is the sum of typed, distance-weighted relation edges to senses of nearby words.

For Mova

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

  • I1 (learn): Run fixed answer/WSD probes against ablated versions of our WordNet/VerbNet knowledge to measure what each relation family actually contributes, using the stripped version as a negative control.
  • I2 (world): Track edge counts per relation type alongside lemma coverage when compiling the ontology, and prioritise filling relation gaps over adding isolated lemmas.
  • I3 (tokens): English derivational pairs (write/writer, teach/teacher) could be linked by lemma-level rules in crate en, giving frames an agent or instrument role for derived nouns.
  • I4 (learn): Use the same pattern + association-score + review pipeline to grow opposition/contrast lexicons that support irony and implicature detection in prag.
  • I5 (frames): This offers an explainable sense-selection step before frame building: every chosen sense can cite the specific relation edges that supported it.

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 MonastirMonastir · Tunisia6 works · 47 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 worksWord sense disambiguation · 83 works

Cite

Cite as Mohamed Ali Batita and Mounir Zrigui. 2019. The Extended Arabic WordNet: a Case Study and an Evaluation using a Word Sense Disambiguation System. Proceedings of the 10th Global Wordnet Conference.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2019.gwc-1.7

@inproceedings{batita2019extended,
  title = {{The Extended Arabic WordNet: a Case Study and an Evaluation using a Word Sense Disambiguation System}},
  author = {Mohamed Ali Batita and Mounir Zrigui},
  booktitle = {Proceedings of the 10th Global Wordnet Conference},
  year = {2019},
  doi = {10.18653/v1/2019.gwc-1.7},
  url = {https://doi.org/10.18653/v1/2019.gwc-1.7}
}

References

44 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Ahmed Abbache, Fatiha Barigou, Fatma Zohra Belkredim, and Ghalem Belalem. 2016. The use of arabic wordnet in arabic information retrieval. In Business Intelligence: Concepts, Methodologies, Tools, and Applications, pages 773–783. IGI Global. ↗
  2. Ahmed Abbache, Farid Meziane, Ghalem Belalem, and Fatma Zohra Belkredim. 2018. Arabic query expansion using wordnet and association rules. In Information Retrieval and Management: Concepts, Methodologies, Tools, and Applications, pages 1239–1254. IGI Global. ↗
  3. Mourad Abbas, Kamel Smaı̈li, and Daoud Berkani. 2011. Evaluation of topic identification methods on arabic corpora. JDIM, 9(5):185–192. find
  4. Ahmed Abdelali, Kareem Darwish, Nadir Durrani, and Hamdy Mubarak. 2016. Farasa: A fast and furious segmenter for arabic. In The Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 11–16.
  5. Lahsen Abouenour, Karim Bouzoubaa, and Paolo Rosso. 2010. Using the yago ontology as a resource for the enrichment of named entities in arabic wordnet. In Proceedings of The Seventh International Conference on Language Resources and Evaluation (LREC 2010) Workshop on Language Resources and Human Language Technology for Semitic Languages, pages 27–31. find
  6. Lahsen Abouenour, Karim Bouzoubaa, and Paolo Rosso. 2013. On the evaluation and improvement of arabic wordnet coverage and usability. Language resources and evaluation, 47(3):891–917. ↗
  7. Musa Alkhalifa and Horacio Rodrı́guez. 2009. Automatically extending ne coverage of arabic wordnet using wikipedia. In Proc. Of the 3rd International Conference on Arabic Language Processing CITALA2009, Rabat, Morocco. find
  8. Ali Alkhatlan, Jugal Kalita, and Ahmed Alhaddad. 2018. Word sense disambiguation for arabic exploiting arabic wordnet and word embedding. Procedia computer science, 142:50–60. find
  9. Fahad Alotaiby, Salah Foda, and Ibrahim Alkharashi. 2014. Arabic vs. english: Comparative statistical study. Arabian Journal for Science and Engineering, 39(2):809–820. find
  10. Mohamed Seghir Hadj Ameur, Ahlem Chérifa Khadir, and Ahmed Guessoum. 2017. An automatic approach for wordnet enrichment applied to arabic wordnet. In International Conference on Arabic Language Processing, pages 3–18. Springer. find
  11. Francesco Barbieri and Horacio Saggion. 2014. Automatic detection of irony and humour in twitter. In ICCC, pages 155–162. find
  12. Mohamed Ali Batita and Mounir Zrigui. 2017. The enrichment of arabic wordnet antonym relations. In International Conference on Computational Linguistics and Intelligent Text Processing, pages 342–353. Springer. find
  13. Mohamed Ali Batita and Mounir Zrigui. 2018. Derivational relations in arabic wordnet. In The 9th Global WordNet Conference GWC, pages 137–144.
  14. William Black, Sabri Elkateb, Horacio Rodriguez, Musa Alkhalifa, Piek Vossen, Adam Pease, and Christiane Fellbaum. 2006. Introducing the arabic wordnet project. In Proceedings of the third international WordNet conference, pages 295–300. Citeseer. find
  15. Mohamed Mahdi Boudabous, Nouha Chaâben Kammoun, Nacef Khedher, Lamia Hadrich Belguith, and Fatiha Sadat. 2013. Arabic wordnet semantic relations enrichment through morpho-lexical patterns. In 2013 1st International Conference on Communications, Signal Processing, and their Applications (ICCSPA), pages 1–6. IEEE. find
  16. Nadia Bouhriz, Faouzia Benabbou, and Habib Benlah-mer. 2015. Text conceptsextraction based on arabic wordnet and formal concept analysis. International Journal of Computer Applications, 111(16):30–34. find
  17. Janez Brank, Marko Grobelnik, and Dunja Mladenic. 2005. A survey of ontology evaluation techniques. In Proceedings of the conference on data mining and data warehouses (SiKDD 2005), pages 166–170. Citeseer Ljubljana, Slovenia. find
  18. Violetta Cavalli-Sforza, Hind Saddiki, Karim Bouzoubaa, Lahsen Abouenour, Mohamed Maamouri, and Emily Goshey. 2013. Bootstrapping a wordnet for an arabic dialect from other wordnets and dictionary resources. In 2013 ACS International Conference on Computer Systems and Applications (AICCSA), pages 1–8. IEEE. find
  19. Udaya Raj Dhungana, Subarna Shakya, Kabita Baral, and Bharat Sharma. 2015. Word sense disambiguation using wsd specific wordnet of polysemy words. In Proceedings of the 2015 IEEE 9th International Conference on Semantic Computing (IEEE ICSC 2015), pages 148–152. IEEE. ↗
  20. Kostas Fragos, Yannis Maistros, and Christos Skourlas. 2003. Word sense disambiguation using wordnet relations. In First Balkan Conference in Informatics, Thessaloniki. find
  21. Gil Francopoulo, Monte George, Nicoletta Calzolari, Monica Monachini, Nuria Bel, Mandy Pet, and Claudia Soria. 2006. Lexical markup framework (lmf). In International Conference on Language Resources and Evaluation-LREC 2006. ↗
  22. Chan-Chia Hsu. 2015. A syntagmatic analysis of antonym co-occurrences in chinese: contrastive constructions and co-occurrence sequences. Corpora, 10(1):47–82. ↗
  23. Farkhund Iqbal, Benjamin CM Fung, Mourad Debbabi, Rabia Batool, and Andrew Marrington. 2019. Wordnet-based criminal networks mining for cybercrime investigation. IEEE Access. ↗
  24. Abdelghani Karkar, Jihad Mohamad Alja’am, Mohamad Eid, and Andrei Sleptchenko. 2015. Elearning mobile application for arabic learners. Journal of Educational & Instructional Studies in the World, 5(2). find
  25. Adam Kilgarriff, Pavel Rychly, Pavel Smrz, and David Tugwell. 2004. Itri-04-08 the sketch engine. Information Technology, 105:116. find
  26. Svetla Koeva, Cvetana Krstev, and Duško Vitas. 2008. Morpho-semantic relations in wordnet–a case study for two slavic languages. In Global wordnet conference, pages 239–253. University of Szeged, Department of Informatics. find
  27. SG Kolte and SG Bhirud. 2009. Wordnet: a knowledge source for word sense disambiguation. International Journal of Recent Trends in Engineering, 2(4). find
  28. Cilia Lachichi, Chahrazad Bendiaf, Lamia Berkani, and Ahmed Guessoum. 2018. An arabic wordnet enrichment approach using machine translation and external linguistic resources. In 2018 2nd International Conference on Natural Language and Speech Processing (ICNLSP), pages 1–6. IEEE. find
  29. Diana McCarthy. 2006. Relating wordnet senses for word sense disambiguation. In Proceedings of the Workshop on Making Sense of Sense: Bringing Psycholinguistics and Computational Linguistics Together. find
  30. Galley Michel and McKeown Kathleen R. 2003. Improving word sense disambiguation in lexical chaining. In Proceedings of the Eighteenth International Joint Conference on Artificial Intelligence, pages 1486–1488. ↗
  31. George A Miller. 1995. Wordnet: a lexical database for english. Communications of the ACM, 38(11):39–41. ↗
  32. George Miller. 1998. WordNet: An electronic lexical database. MIT press. find
  33. Verginica Barbu Mititelu. 2012. Adding morphosemantic relations to the romanian wordnet. In LREC, pages 2596–2601. find
  34. Steven Neale. 2018. A survey on automatically-constructed wordnets and their evaluation: Lexical and word embedding-based approaches. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC-2018). find
  35. Yasser Regragui, Lahsen Abouenour, Fettoum Krieche, Karim Bouzoubaa, and Paolo Rosso. 2016. Arabic wordnet: New content and new applications. In Proceedings of the Eighth Global WordNet Conference, pages 330–338.
  36. Horacio Rodrı́guez, David Farwell, Javi Farreres, Manuel Bertran, Musa Alkhalifa, M Antonia Martı́, William Black, Sabri Elkateb, James Kirk, Adam Pease, et al. 2008. Arabic wordnet: Current state and future extensions. In Proceedings of The Fourth Global WordNet Conference, Szeged, Hungary, pages 387–405. find
  37. Abdulgabbar Saif, Mohd Juzaiddin Ab Aziz, and Nazlia Omar. 2017. Mapping arabic wordnet synsets to wikipedia articles using monolingual and bilingual features. Natural Language Engineering, 23(1):53–91. find
  38. Hee-Cheol Seo, Hoojung Chung, Hae-Chang Rim, Sung Hyon Myaeng, and Soo-Hong Kim. 2004. Unsupervised word sense disambiguation using wordnet relatives. Computer Speech & Language, 18(3):253–273. ↗
  39. Advaith Siddharthan, Nicolas Cherbuin, Paul J Eslinger, Kasia Kozlowska, Nora A Murphy, and Leroy Lowe. 2018. Wordnet-feelings: A linguistic categorisation of human feelings. arXiv preprint arXiv:1811.02435. ↗
  40. Krešimir Šojat, Matea Srebačić, and Marko Tadić. 2012. Derivational and semantic relations of croatian verbs. Journal of Language Modelling, pages 111–142. ↗
  41. Nasrin Taghizadeh and Hesham Faili. 2016. Automatic wordnet development for low-resource languages using cross-lingual wsd. Journal of Artificial Intelligence Research, 56:61–87. ↗
  42. Piek Vossen. 1998. A multilingual database with lexical semantic networks. Dordrecht: Kluwer Academic Publishers. doi, 10:978–94. find
  43. A Zouaghi, L Merhbene, and M Zrigui. 2011. Word sense disambiguation for arabic language using the variants of the lesk algorithm. WORLDCOMP, 11:561–567. find
  44. Anis Zouaghi, Laroussi Merhbene, and Mounir Zrigui. 2012. Combination of information retrieval methods with lesk algorithm for arabic word sense disambiguation. Artificial Intelligence Review, 38(4):257– 269. ↗

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/batita-2019-extended-arabic-wordnet-case-study/