A state of the art of natural language processing of the Tunisian dialect

In the area of Arabic Natural Language Processing, most of the undertaken research and achievements have mainly involved Modern Standard Arabic (MSA). The various Arabic dialects (AD) are still considered to be among under-resourced languages. It's only in the last decade that these dialects began to arouse the interest of NLP researchers, especially given their increasing use on the social web. In this work, we focus on the Tunisian dialect (TD), and propose to provide a state of the art on the automatic processing of this dialect. A review of the works carried out to date and a detailed inventory of the various NLP tools and language resources available for the TD are presented and discussed.