Predicting Dialect Variation in Immigrant Contexts Using Light Verb Constructions

Abstract

Languages spoken by immigrants change due to contact with the local languages. Capturing these changes is problematic for current language technologies, which are typically developed for speakers of the standard dialect only. Even when dialectal variants are available for such technologies, we still need to predict which dialect is being used. In this study, we distinguish between the immigrant and the standard dialect of Turkish by focusing on Light Verb Constructions. We experiment with a number of grammatical and contextual features, achieving over 84% accuracy (56% baseline).

Computational Sociolinguistics: A Survey

“…Whereas some have considered automatic language identification to be a solved problem (McNamee 2005), many outstanding issues still exist (Hughes et al 2006), including the identification of dialects and closely related languages (Zampieri et al 2014(Zampieri et al , 2015. In studies on automatic dialect identification, various dialects have been explored, including Arabic (Elfardy and Diab 2013;Zaidan and Callison-Burch 2013;Darwish, Sajjad, and Mubarak 2014;Huang 2015), Turkish (Dogruöz and Nakov 2014), Swiss German (Scherrer and Rambow 2010), and Dutch (Trieschnigg et al 2012) dialects.…”

“…Their method is based on a linguistic atlas for the extraction of lexical, morphological, and phonetic rules and the likelihood of these forms across German-speaking Switzerland. Dogruöz and Nakov (2014) explored the use of light verb constructions to distinguish between two Turkish dialects.…”