A description and demonstration of SAFAR framework
Karim Bouzoubaa1,
Younes Jaafar1, Driss Namly1, Ridouane Tachikart1,
Rachida Tajmout1,
Hakima Khamar2, Hamid Jaafar3, Lhoussain Aouragh4,
Abdellah Yousfi5
1 Mohammadia School
of Engineers, Mohammed V University in Rabat, Morocco
2 Faculty of
Letters and Human Sciences, Mohammed Vth University, Rabat, Morocco
3 Polidisciplinary
faculty of Safi, Caddi Ayyad University, Morocco
4 Faculty of Legal,
Economic and Social Sciences – Sale, Mohammed V University in Rabat, Morocco.
5 Faculty of Legal,
Economic and Social Sciences – Souissi, Mohammed V University in Rabat,
Morocco.
karim.bouzoubaa@emi.ac.ma; jayounes@yahoo.fr; namly_driss@yahoo.fr;
tachicart@gmail.com; tajmoutrachida@yahoo.fr; khamarhaki@gmail.com;
jaafarhamid1973@gmail.com; jaafarhamid1973@gmail.com; aouragh@hotmail.com;
yousfi240ma@yahoo.fr
Abstract
Several tools and resources has been developed
to deal with Arabic NLP problems. However, a homogenous and flexible Arabic
environment that gathers these resources is rarely available. In this
perspective, we introduce SAFAR framework which is a monolingual framework
dedicated to Arabic language. It is considered as a repository and collaborative
work where multiple developers of Arabic tools and resources can meet and share
their products. During two decades
SAFAR integrates more than 50 tools and resources. Moreover, a benchmark with
similar frameworks dealing with Arabic language shows a prevalence of SAFAR
regarding different features such as maintenance or software engineering
principles
Keywords: NLP architecture, NLP infrastructure, framework, Arabic language, standardization, Software Engineering
Next section