On two SweLL learner corpora – SweLL-pilot and SweLL-gold

Authors

  • Elena Volodina

DOI:

https://doi.org/10.3384/ecp205012

Keywords:

SweLL, learner corpus research infrastructure, Swedish as a second language, correction annotation aka error annotation, normalization, CEFR labels

Abstract

SweLL – Swedish Learner Language – is a unifying term for the infrastructure module for research on Swedish as a Second Language (L2), deployed and maintained as a part of bigger infrastructure of Språkbanken Text at the University of Gothenburg, Sweden. The SweLL infrastructure module consists of a number of learner data collections, and tools for annotation and management of learner data. As a result, many of its components contain the prefix SweLL in their names, which has created some confusion, especially with regards to the two corpora. In this article we shortly introduce the various SweLL-components with a special focus on the differences between the two SweLL corpora.

Downloads

Published

2024-01-04