The Parsing of Simple Arabic Verbal Sentences Using NooJ Platform

Abstract

In this paper, we present a NooJ parser of simple Arabic verbal sentence. This parser is based on dependency grammar established by the attribution ( Open image in new window , al-’isnād) concept in the Arabic language. In the first part of this paper, we present a syntactic and semantic classification of Arabic words allowing Arabic sentence parsing. In the second part, we present the shared structure by simple Arabic sentences. Furthermore, we use this structure to implement simple Arabic verbal sentence grammar using NooJ platform. Our parser is applied to the input sentence after two required steps: Morphological analysis and morphological disambiguation. The proposed parser generates possible parse tree(s) of the input sentence, and annotates all sentence components by their grammatical functions. The implemented parser is tested on a selected text; experimental results show its efficacy.