Derivational Relations in Arabic WordNet

Mohamed Ali Batita, Mounir Zrigui


Abstract
When derivational relations deficiency exists in a wordnet, such as the Arabic WordNet, it makes it very difficult to exploit in the natural language processing community. Such deficiency is raised when many wordnets follow the same development path of Princeton WordNet. A rule-based approach for Arabic derivational relations is proposed in this paper to deal with this deficiency. The proposed approach is explained step by step. It involves the gathering of lexical entries that share the same root, into a bag of words. Rules are then used to affect the appropriate derivational relations, i.e. to relate existing words in the AWN, involving part-of-speech switch. The method is implemented using Java. Manual verification by a lexicographer takes place to ensure good results. The described approach gave good results. It could be useful for other morphologically complex languages as well.
Anthology ID:
2018.gwc-1.16
Volume:
Proceedings of the 9th Global Wordnet Conference
Month:
January
Year:
2018
Address:
Nanyang Technological University (NTU), Singapore
Editors:
Francis Bond, Piek Vossen, Christiane Fellbaum
Venue:
GWC
SIG:
SIGLEX
Publisher:
Global Wordnet Association
Note:
Pages:
136–144
Language:
URL:
https://aclanthology.org/2018.gwc-1.16
DOI:
Bibkey:
Cite (ACL):
Mohamed Ali Batita and Mounir Zrigui. 2018. Derivational Relations in Arabic WordNet. In Proceedings of the 9th Global Wordnet Conference, pages 136–144, Nanyang Technological University (NTU), Singapore. Global Wordnet Association.
Cite (Informal):
Derivational Relations in Arabic WordNet (Batita & Zrigui, GWC 2018)
Copy Citation:
PDF:
https://aclanthology.org/2018.gwc-1.16.pdf