Recent decades have seen an increase in the prevalence of the use of computational methods in
the study of language, including in sociolinguistics. These methods allow for the study of
language variation through the analysis of social media data and even for the mapping of the
spread of linguistic variation in the real world. The goal of this study was to assess the utility
of computational methods in the extraction of Dutch Low Saxon dialect features from a large
Twitter corpus. The results ...
Recent decades have seen an increase in the prevalence of the use of computational methods in
the study of language, including in sociolinguistics. These methods allow for the study of
language variation through the analysis of social media data and even for the mapping of the
spread of linguistic variation in the real world. The goal of this study was to assess the utility
of computational methods in the extraction of Dutch Low Saxon dialect features from a large
Twitter corpus. The results indicate that these dialect features can be used successfully for the
training of classifiers and that maps generated based on these features and their associated
predictions have the potential to capture the use of Dutch Low Saxon in the Netherlands,
although methodological adjustments are advisable for future studies. The study of
socioeconomic status as it relates to Low Saxon proved feasible, although to a limited degree.
+
