Elamur Research⌕
☰
home / papers / 2014 / Azhary: An Arabic Lexical Ontology

Azhary: An Arabic Lexical Ontology

Summary

The authors argue that Arabic, despite having hundreds of millions of speakers and great religious and cultural weight, lacks a dependable semantic lexicon for computational work. They present Azhary, a WordNet-style lexical ontology for Arabic. Like WordNet, it gathers words into synonym groups (synsets) and links them by typed lexical relations: synonymy, antonymy, hypernymy/hyponymy, meronymy/holonymy, and an extra "association" relation. The final resource holds 26,195 words in 13,328 synsets. The authors built it with Arabic Wordnet (AWN), the most widely used existing Arabic lexical ontology, as the point of comparison. They report that Azhary covers more words, stores more relations between words, and answers lookups faster than AWN. The paper also surveys earlier Arabic ontology work in Quranic, linguistic, automatically generated and domain-specific settings, and notes that most of it is small, narrow, or unfinished. The contribution matters because it supplies a broader general-purpose semantic backbone for Arabic NLP tasks such as disambiguation, text understanding and machine translation.

Key points

  • Azhary contains 26,195 words grouped into 13,328 synsets. AWN, by comparison, has 11,270 synsets and 23,496 expressions (words and multiwords).
  • The relation inventory covers synonym, antonym, hypernym, hyponym, meronym and holonym, plus an "association" link that goes beyond the classic WordNet set.
  • The evaluation is a direct comparison with AWN, in which Azhary comes out ahead on vocabulary size, number of relations and response time.
  • The reference list cites Protégé, Apache Jena and JExcelApi as tools, and several Arabic dictionaries (almaany, the Mediator Lexicon, a "rich lexicon") as lexical sources.
  • The related-work survey highlights small, limited predecessors. Examples include a Quranic time-noun ontology with 18 classes and 59 words, and an upper-level Arabic ontology with only 420 evaluated concepts.

Interesting ideas

  • I1. Adding an explicit "association" relation alongside the taxonomic and part-whole links captures loose topical relatedness without forcing it into an is-a or part-of slot. — For Mova (world): Our WordNet-derived ontology could carry a separate, clearly labelled association edge. Reasoning could then use topical links only as weak evidence and keep them out of entailment chains.
  • I2. The authors criticise ontologies built purely from shared roots: words derived from one root may share a core sense yet belong to different semantic categories. — For Mova (tokens): Lemma and derivational-morphology links (e.g., "employ"/"employee"/"employment") should not be treated as proof of semantic class membership. Class should come from the ontology, with morphology used only as a hint.
  • I3. The new resource is judged against the dominant existing one on concrete, comparable axes: coverage, relation count and lookup latency. — For Mova (learn): When we compile a new knowledge layer, we can routinely benchmark it against the WordNet/Wikidata baseline on the same measurable axes. That turns "the new layer is better" into a checkable claim.
  • I4. The survey shows that many small domain ontologies (Quran, legal, blogs) never reached general use, which argues for one broad lexical core with domain extensions attached. — For Mova (world): Domain vocabularies, such as the measures and values for maths word problems, should hang off the shared ontology rather than live in separate islands. This keeps the system explainable across texts.

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

Fields: Linguistics & languages

Cited by

3 papers of our library cite it (of the 140884 whose bibliographies we read); 1 of them have a page here, 2 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.1411.1999 · arXiv 1411.1999

@misc{ishkewy2014azhary,
  title = {{Azhary: An Arabic Lexical Ontology}},
  author = {Ishkewy, Hossam and Harb, Hany and Farahat, Hassan},
  year = {2014},
  doi = {10.48550/arxiv.1411.1999},
  eprint = {1411.1999},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.1411.1999}
}

References

33 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Farhan El-Salim,"خاــغللا يـيت اـهرًاكهو حــيترـعلا حــغللا ", Available: http://www.saaid.net/Minute/33.htm, [Accessed: Aug. 5, 2014]. ↗
  2. Haytham T. Al-Feel, Magdy Koutb and Hoda Suoror, "Semantic Web on Scope: A New Architectural Model for the Semantic Web",Journal of Computer Science 4(7): 613-624, 2008. find
  3. Tim Berners-Lee, James Hendler and Ora Lassila,"The Semantic Web", Scientific American, May 17, 2001 find
  4. T. Berners-Lee, "Artificial Intelligence & the Semantic Web" World Wide Web Consortium, 2006. [Online]. Available: http://www.w3.org/2006/Talks/0718-aaai-tbl/Overview.html#(14) [Accessed: Sep. 5, 2014]. ↗
  5. Liyang Yu, "Introduction to the Semantic Web and Semantic Web Services",2007. find
  6. T. R. Gruber, "A Translation Approach to a Portable Ontology Specification", Knowledge Acquisition, vol. 6, pp. 199-221, 1993. find
  7. G. A. Miller, R. Beckwith, C. Fellbaum, D. Gross, and K. J. Miller " Introduction to WordNet: an on-line lexical database ", International Journal of Lexicography 3(4):235-244 (1990). find
  8. http://en.wikipedia.org/wiki/WordNet [Accessed: Sep. 9, 2014]. ↗
  9. Aya M. Al‐Zoghbya , Ahmed Sharaf Eldin Ahmed, Taher T. Hamza, "Arabic Semantic Web Applications – A Survey", Semantic Web Journal, 2012. find
  10. http://globalwordnet.org/wordnets-in-the-world[Accessed: Sep. 7, 2014]. ↗
  11. Christiane Fellbaum, Musa Alkhalifa, William J.Black, Sabri Elkateb, Adam Pease, HoracioRodríguez, and Piek Vossen, "Building a WordNet for Arabic", May, 2006. AWN 1.3 Second 23,496 11,270 Yes Yes Yes No No No No Azhary 11 Millie Second 26,195 13,328 Yes Yes Yes Yes Yes Yes Yes find
  12. Maha Al-Yahya, Hend Al-Khalifa,Alia Bahanshal, Iman Al-Odah, Nawal Al Helwah, "AN ONTOLOGICAL MODEL FOR REPRESENTING SEMANTIC LEXICONS: AN APPLICATION ON TIME NOUNS IN THE HOLY QURAN", The Arabian Journal for Science and Engineering, Volume 35, December 2010. find
  13. A. De Nicola, M. Missikoff, and R. Navigli, “A Software Engineering Approach to Ontology Building”,Information Systems, 34(2009), pp. 258–275. find
  14. Hikmat Ullah Khan, Syed Muhammad Saqlain, Muhammad Shoaib,Muhammad Sher, "Ontology Based Semantic Search in Holy Quran", International Journal of Future Computer and Communication, Vol. 2, No. 6, December, 2013. find
  15. Azman Ta’a, Syuhada Zainal Abidin, Mohd Syazwan Abdullah, Abdul Bashah B Mat Ali, and Muhammad Ahmad," AL-QURAN THEMES CLASSIFICATION USING ONTOLOGY", Proceedings of the 4the International Conference on Computing and Informatics, ICOCI, 2013. find
  16. Mohamed Ahmed Sherif ,Axel-Cyrille Ngonga Ngomo, "Semantic Quran: A Multilingual Resource for Natural-Language Processing", Semantic Web Journal,2009 find
  17. Fatma Zohra Belkredim, Ali El Sebai "An Ontology Based Formalism for the Arabic Language Using Verbs and their Derivatives", IBIMA,Volume 11, 2009. find
  18. Mustafa Jarrar, "Building a Formal Arabic Ontology" , In proceedings of the Experts Meeting on Arabic Ontologies and Semantic Networks April 26-28, 2011. find
  19. Hassina Aliane, Zaia Alimazighi, Mazari Ahmed Cherif,"Al–Khalil: The Arabic Linguistic Ontology Project", Seventh International Conference on Language Resources and Evaluation, 2010. find
  20. Musa Alkhalifa, Horacio Rodríguez, "Automatically Extending Named Entities Coverage of Arabic WordNet using Wikipedia ",International Journal on Information and Communication Technologies, Vol. 3, No. 3, June 2010. find
  21. Lahsen Abouenour," On the Improvement of Passage Retrieval in Arabic Question/Answering (Q/A) Systems", Natural Language Processing and Information Systems, Volume 6716, 2011, pp 336-341. find
  22. Abdurrahman A. Nasr., M.Sc. thesis, Al-Azhar university, Faculty Of Engineering, Computers and Systems Dept., 2012. find
  23. Rodríguez, H. Farwell, D. Farreres, J. Bertran, M. Alkhalifa, M. Martí M.A " Arabic WordNet: Semi-automatic Extensions using Bayesian Inference". In Proceedings of the the 6th Conference on Language Resources and Evaluation LREC,May, 2008. find
  24. Lahsen Abouenour, Karim Bouzoubaa, Paolo Rosso," Using the Yago ontology as a resource for the enrichment of Named Entities in Arabic WordNet ", Workshop on LR & HLT for Semitic Languages, 7th Int. Conf. on Language Resources and Evaluation, LREC, May, 2010. find
  25. Saidah Saad, Naomie Salim and Hakim Zainal,"Islamic Knowledge Ontology Creation", Internet Technology and Secured Transactions, 2009. find
  26. Ahmed Cherif Mazari , Hassina Aliane, and Zaia Alimazighi ,"Automatic construction of ontology from Arabic texts" ,Proceedings ICWIT, 2012. find
  27. Saidah Saad, Naomi Salim, "Methodology of Ontology Extraction for Islamic Knowledge Text", In Postgraduate Annual Research Seminar, UTM,2008. find
  28. Samhaa R. El-Beltagy, Maryam Hazman, Ahmed Rafea, "Ontology learning from domain specific web documents," vol. 4, No. 1/2, May 2009. find
  29. S. Zaidi, M.T. Laskri, K. Bechkoum, "A Cross-language Information Retrieval Based on an Arabic Ontology in the Legal Domain, IEEE SITIS" ,2005. find
  30. Lilac Al-Safadi, Mai Al-Badrani, Meshael Al-Junidey, "Developing Ontology for Arabic Blogs Retrieval", International Journal of Computer Applications (0975– 8887),Volume 19– No.4, April, 2011. find
  31. Ibrahim Fathy Moawad, Mohammad Abdeen, Mostafa Mahmoud Aref, "Ontology-based Architecture for an Arabic Semantic Search Engine", December 15-16, 2010. find
  32. Abd El-Ghany Abo El-Azm, rich lexicon(يٌغلا نجعه),2013 find
  33. Fadel Mostafa El-Saki, "Arab parts of speech in terms of form and function" (حفيظىلاو لكشلا ثيح يه ًترعلا ملاكلا ماسقأ),1977. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/ishkewy-2014-azhary-arabic-lexical-ontology/