Creating a Lexicon of Bavarian Dialect by Means of Facebook Language Data and Crowdsourcing

Creating a Lexicon of Bavarian Dialect by Means of Facebook Language Data and Crowdsourcing Manuel Burghardt, Daniel Granvogl, Christian Wolff Media Informatics Group Institute for Information and Media, Language and Culture University of Regensburg, Germany Email: manuel.burghardt@ur.de, daniel@granvogl.com, christian.wolff@ur.de Abstract Data acquisition in dialectology is typically a tedious task, as dialect samples of spoken language have to be collected via questionnaires or interviews. In this article, we suggest to use the “web as a corpus” approach for dialectology. We present a case study that demonstrates how authentic language data for the Bavarian dialect (ISO 639-3:bar) can be collected automatically from the social network Facebook. We also show that Facebook can be used effectively as a crowdsourcing platform, where users are willing to translate dialect words collaboratively in order to create a common lexicon of their Bavarian dialect. Key insights from the case study are summarized as “lessons learned”, together with suggestions for future enhancements of the lexicon creation approach. Keywords: dialectology, Bavarian, ISO 639-3:bar, dialect lexicon, crowdsourcing, social media, Facebook 1. Introduction: Dialectology and the Internet Dialectology is a branch of sociolinguistics that typically examines instances of spoken language. This makes data acquisition a tedious task, as dialect samples of spoken language have to be collected via questionnaires or interviews. Furthermore, data collected in such a way needs to be transcribed and normalized. In corpus linguistics, there has been a trend to utilize the “web as a corpus” during the last years (Kilgarriff & Grefenstette, 2003; Baroni et al., 2009). More recently, there are also a number of studies from the field of dialectology that try to make use of language resources available from the Internet: For the case of Swiss dialect, Siebenhaar (2003) estimates that around 22% percent of Swiss websites contain text written in dialect. In a consecutive study on Swiss Internet Relay Chats (IRC), Siebenhaar (2005) finds that 80-90% of the messages posted in the chatrooms he analyzed are written dialect. Ziegler (2005) also presents a study on IRC chats, studying how German dialect is realized in chatrooms, and how it differs from standard language use. As IRC is quickly becoming outdated in the light of more recent social media platforms such as Twitter or Facebook, we believe that dialectologists should also try to make use of language data created by the users of these novel social networks. In this article, we present a case study that demonstrates how authentic language data for a Bavarian dialect (ISO 639-3:bar) can be collected automatically from the social network Facebook. We also show that Facebook can be used effectively as a crowdsourcing platform, where users are willing to collaboratively translate dialect words in order to create a common lexicon of their Bavarian dialect. 2. Overview of Bavarian Dialect Bavaria is one of 16 federal states in Germany. However, it is important to distinguish the state of Bavaria (“Freistaat Bayern”) and the Bavarian dialect, which is not per se identical: Not all inhabitants of Bavaria speak Bavarian dialect, and there are …

Content truncated to 500 words