A linguistic method for Arabic normative provisions' annotation based on contextual exploration

Abstract

We present in this paper a linguistic method to automatically identify and semantically annotate the categories of normative provisions in Arabic legal texts. The proposed method is based on a contextual exploration of the discursive knowledge conveyed in the processed texts. Coupled to a taxonomy of Arabic normative provisions and a set of annotation rules, this method enables the semantic annotation of the normative provisions based on a surface analysis in order to extract a set of relevant linguistic markers having contextual dependencies. The performance of this method has been evaluated by developing a prototype to automatically identify the normative categories over a large set of Arabic normative texts collected from the Official Gazette of the Republic of Tunisia. Evaluation was conducted in terms of Precision, Recall and F-score. We obtained respectively 96,4%, 96,06% and 96,23% for Precision, Recall and F-score.