A Turkish-German Code-Switching Corpus

A Turkish-German Code-Switching Corpus Özlem Çetino˘glu IMS, University of Stuttgart Germany ozlem@ims.uni-stuttgart.de Abstract Bilingual communities often alternate between languages both in spoken and written communication. One such community, Germany residents of Turkish origin produce Turkish-German code-switching, by heavily mixing two languages at discourse, sentence, or word level. Code-switching in general, and Turkish-German code-switching in particular, has been studied for a long time from a linguistic perspective. Yet resources to study them from a more computational perspective are limited due to either small size or licence issues. In this work we contribute the solution of this problem with a corpus. We present a Turkish-German code-switching corpus which consists of 1029 tweets, with a majority of intra-sentential switches. We share different type of code-switching we have observed in our collection and describe our processing steps. The first step is data collection and filtering. This is followed by manual tokenisation and normalisation. And finally, we annotate data with word-level language identification information. The resulting corpus is available for research purposes. Keywords: code-switching, Turkish, German 1. Introduction Code-switching (CS) is a natural extension of the language used among immigrant communities (Toribio and Bullock, 2012). Bilingual speakers fluently switch between the lan- guage of their background culture and the language of the country they live in, by alternating inter-sentence, intra- sentence, or even intra-word. This is especially easy to ob- serve in daily life where one immigrant group constitutes a large group of minority such as the Turkish community in Germany. Example 1 gives such a code-switching in- stance:1 (1) seitdem ich keine schule hab cok faul oldum ben ya:D ‘since I have no school I became very lazy :D’ Due to the large number of Turkish immigrants in Germany, Turkish-German CS is studied by several researchers, mostly on the grounds of sociolinguistics and language ac- quisition. Kallmeyer and Keim (2003) investigate the char- acter of communication between young girls in Mannheim, mostly of Turkish origin and show, among other things, that they employ a mixed form of Turkish and German with peers. Androutsopoulos and Hinnenkamp (2001) and Hinnenkamp (2008) look into chat rooms of immigrants of Greek and Turkish origin to observe the multilingual be- haviour. The SKOBI Corpus collects the recordings of Turkish- German bilingual children (Rehbein et al., 2009a). Rehbein et. al (2009b) and Herkenrath (2012), who primarily study the development of the complex language these children employ, report on CS in their language. Kiezdeutsch cor- pus (Rehbein et al., 2014) consists of conversations among native German adolescents from a multiethnic urban area. Since the majority of the participants have Turkish back- ground, about 400 Turkish sentences found place in the cor- pus, 40 of which employ intra-sentential CS. 1German parts are in bold. Although code-switching is more observed in spoken data, written data also exhibits CS, especially in the informal, conversation-like environment of social media. User gen- erated content is as interesting as data collected from field work. However, its large size raises the need of computational approaches to code-switching. Researchers …

Content truncated to 500 words