Arabic Probabilistic Context Free Grammar Induction from a Treebank

Linguistic resources are very important to any natural language processing task. Unfortunately, the manual construction of these resources is laborious and time-consuming. The use of annotated corpora as a knowledge database might be a solution to a fast construction of a grammar for a given language. In this paper, we present our method to automatically induce a syntactic grammar from an Arabic annotated corpus (The Penn Arabic TreeBank), a probabilistic context free grammar in our case. To construct our resource, we first induce context free rules from the annotated corpus trees as a first step and then we calculate a specific probability for each induced rule. Finally, we present and discuss the obtained grammar.