wa3dbk/Barcha

Open source NLP resources for the Tunisian arabic dialect. # Barcha Open source speech and natural language processing resources for the Tunisian arabic dialect (work in progress). ------------------- ## Ressources The data and ressources collected within this project is multi-purpose ; named entities recognition, machine translation, language modelling, .. ### Named entities List of named entities : #### People - List of Tunisian academics and scientists - List of Tunisian artists - List of Tunisian football players - List of Tunisian media personalities (presenters, directors, producers, ..) - List of Tunisian ministries - List of Tunisian poets - List of Tunisian politicians - List of Tunisian trade unionists - List of Tunisian writers #### Institutions, associations and companies - List of Tunisian associations - List of Tunisian companies - List of Tunisian governmental institutions #### Places - List of Tunisian cities #### Institutions - List of Tunisian Higher Institutes for Technological Studies (ISET) - List of Tunisian political parties - List of Tunisian Universities (private) - List of Tunisian Universities (public) - List of Tunisian Unions ### Todo : - [ ] Collect more raw text data in Tunisian arabic. - [ ] Develop cleaning / spelling correction scripts for Tunisian arabic. - [ ] Develop CODA-compatible normalization scripts for Tunisian arabic. - [ ] Develop Arabizi / arabic conversion scripts. - [ ] Develop scrapers for Tunisian news/forums websites. - [ ] Build parallel datasets for machine translation between Tunisian english / MSA. - [ ] Develop translation systems for Tunisian English and Tunisian MSA. ## References **CODA:** Habash, Nizar, Mona T. Diab, and Owen Rambow. *"Conventional Orthography for Dialectal Arabic."* LREC. 2012. Zribi, Inès, et al. *"A Conventional Orthography for Tunisian Arabic.*" LREC. 2014. Turki, Houcemeddine, et al. *"A conventional orthography for maghrebi arabic."* Proceedings of the International Conference on Language Resources and Evaluation (L …