The SMarT Classifier for Arabic Fine-Grained Dialect Identification

Karima Meftouh (1, 2) , Karima Abidi (2) , Salima Harrat (2, 3) , Kamel Smaïli (2)

1 LRI-ANNABA - Laboratoire de Recherche en Informatique
2 SMarT - Statistical Machine Translation and Speech Modelization and Text
3 ENS Bouzaréah-Alger - École normale supérieure - Bouzaréah-Alger

This paper describes the approach adopted by the SMarT research group to build a dialect identification system in the framework of the Madar shared task on Arabic fine-grained dialect identification. We experimented several approaches, but we finally decided to use a Multinomial Naïve Bayes classifier based on word and character ngrams in addition to the language model probabilities. We achieved a score of 67.73% in terms of Macro accuracy and a macro-averaged F1-score of 67.31%.

Mots clés: Dialect identification; Multinomial Naïve Bayes; Arabic dialects; Algerian dialect

Domaines: Informatique et langage [cs.CL]

Karima Meftouh, Karima Abidi, Salima Harrat, Kamel Smaïli. The SMarT Classifier for Arabic Fine-Grained Dialect Identification. The Fourth Arabic Natural Language Processing Workshop co-located with ACL, Aug 2019, Florence, Italy. ⟨hal-02166384⟩