N-gram Language Models and POS Distribution for the Identification of Spanish Varieties (Ngrammes et Traits Morphosyntaxiques pour la Identification de Variétés de l'Espagnol) [in French]

This article presents supervised computational methods for the identification of Spanish varieties. The features used for this task were the classical character and word n-gram language models as well as POS and morphological information. The use of these features is to our knowledge new and we aim to explore the extent to which it is possible to identify language varieties solely based on grammatical differences. Four journalistic corpora from different countries were used in these experiments : Spain, Argentina, Mexico and Peru. MOTS-CLÉS : classification automatique, ngrammes, espagnol, variétés nationales.