Speakers: Monojit Choudhury and Kalika Bali

Multilingual communities exhibit code-switching, that is, mixing of two or more socially stable languages in a single conversation, sometimes even in a single utterance. This phenomenon has been widely studied by linguists and interaction scientists in the spoken language of such communities. However, with the prevalence of social media and other informal interactive platforms, code-switching is now also ubiquitously observed in user-generated text. As multilingual communities are more the norm from a global perspective, it becomes essential that code-switched text and speech are adequately handled by language technologies and NUIs like Skype Translator and Cortana.

Project Mélange aims to analyze and understand code-switching behavior at two levels: first, the formal structural level that dictates the grammar of such a construct, and second, the functional level that motivates its use from a cognitive, pragmatic and socio-cultural perspective. This in turn would allow us to process mixed language as well as better model conversations and dialogues in a multilingual setting.

In this talk we will present an overview of our work towards making this a possibility. In particular, we will discuss word-level language detection, and linguistic analysis and automatic identification of functions of switching in Twitter.