On Friday, June 19 there will be a talk by Dong Nguyen (University of Twente). The talks is entitled “Regional Dialects on Twitter”. Abstract: Social media data is increasingly used to study the social dynamics of language. In this talk I will present our research on dialects in the Netherlands based on Twitter data. While the usage of some words varies markedly by geographic region, such as whether ‘patat’ or ‘friet’ (‘french fries’) is used, other patterns are less clear. Quantifying the degree of spatial dependence for linguistic features is a key task for analyzing dialectal variation. I will present a new test that builds on kernel-based methods for non-parametric statistics and discuss experiments on both synthetic data and geotagged tweets from the Netherlands.
Dong Nguyen is a PhD student at the University of Twente. She is also affiliated with the Meertens Institute. She is interested in developing text mining methods that can help answer questions from the social sciences and the humanities. She especially enjoys working with social media data.