TRAVOLTA - Tracing Attitudes And Variation In Online Luxembourgish Text Archives - FHSE

The project at a glance

Start date:

01 Sep 2023
Duration in months:

36
Funding:

Luxembourg National Research Fund (FNR)
Principal Investigator(s):

Christoph PURSCHKE

About

TRAVOLTA combines Natural Language Processing with sociolinguistic analysis to trace the development of written Luxembourgish & public discourse online. In recent years, Luxembourgish has been in the focus of public attention. Its role as the national language of Luxembourg and important means of integration has been widely discussed in the context of a highly multilingual society. On top of that, it has recently developed into a fully fledged written language, largely driven by the rise of social media. However, there is still hardly any research that tackles the societal upswing of Luxembourgish and documents the emergence of a new written language in the digital age. The TRAVOLTA project closes this gap by taking a close look at online text archives of written Luxembourgish from RTL.lu (news items, user comments). One the one hand, it analyses the development of a shared writing practice in digital media as an example for how a new standard language emerges. On the other hand, it monitors the development of public discourse by investigating writers’ attitudes toward pertinent topics such as multilingualism and the COVID-19 pandemic. In doing so, the project also advances the development of computational tools for Luxembourgish as a lesser-resourced language. It provides new solutions for processing text data, that is, automatic spelling correction, variant classification and the extraction of social meaning from text. These tools will be beneficial beyond the scope of the project, for example, in the context of machine translation and human-machine-interaction.