Title: Exploiting the Arabic Wikipedia for semi-automatic construction of a lexical ontology

Authors: Mohamed Mahdi Boudabous; Lamia Hadrich Belguith; Fatiha Sadat

Addresses: ANLP Research Group, MIRACL Laboratory, Faculty of Economics and Management of Sfax, Sfax, Tunisia ' ANLP Research Group, MIRACL Laboratory, Faculty of Economics and Management of Sfax, Sfax, Tunisia ' Department of Computer Science, University of Quebec in Montreal (UQAM), Montreal, Canada

Abstract: In this paper, we propose a hybrid (numerical/linguistic) method to build a lexical ontology for the Arabic language. This method is based on the Arabic Wikipedia. It consists of two phases: analysing the description section in order to build core ontology and then using the physical structure of Wikipedia articles (info-boxes, category pages and redirect links) and their contents for enriching the core ontology. The building phase of the core ontology is implemented via the TBAO system. The obtained core ontology contains more than 200,000 concepts.

Keywords: Arabic Wikipedia; Arabic ontology; lexical ontology; morpho-lexical patterns; semantic relations; semi-automatic construction.

DOI: 10.1504/IJMSO.2013.057768

International Journal of Metadata, Semantics and Ontologies, 2013 Vol.8 No.3, pp.245 - 253

Received: 31 Jan 2013
Accepted: 26 Jun 2013

Published online: 14 Oct 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article