HTTPS SSH

Discriminating between Similar Languages

The Discriminating between Similar Languages (DSL) shared task aims to provide a dataset to evaluate system's performance on discriminating 13 different languages in 6 groups of languages:

  • Group A (Bosnian, Croatian, Serbian)
  • Group B (Indonesian, Malaysian)
  • Group C (Czech, Slovakian)
  • Group D (Brazilian Portuguese, European Portuguese)
  • Group E (Peninsular Spain, Argentine Spanish)
  • Group F (American English, British English)

Shared task webpage: http://corporavm.uni-koeln.de/vardial/sharedtask.html

For reproducibility,

Participants

22 teams registered for the shared task and 7 teams submitted systems:

  • CLCG: Johannes Bjerva - University of Groningen
  • LIRA: Tiina Puolakainen - University of Tartu
  • NRC-CNRC: Cyril Goutte, Serge Leger and Marine Carpuat - National Research Council Canada
  • QMUL: Matthew Purver - Queen Mary University of London
  • RAE: Jordi Porta and José Luis Sancho - Centro de Estudios de la Real Academia Española
  • UDE: Torsten Zesch - University of Duisburg-Essen
  • UMich: Ben King and Steven Abney - University of Michigan
  • UniMelb-NLP: Marco Lui, Ned Letcher, Tim Baldwin, Paul Cook, Li Wang, Oliver Adams and Duong Thanh Long - University of Melbourne / NICTA VRL

Results

The overall results for the DSL shared task can be viewed here

The results for the individual teams can be found here

Detailed results can be viewed here

Organizers

  • Marcos Zampieri, Saarland University
  • Liling Tan, Saarland University
  • Nikola Ljubešić, University of Zagreb
  • Jörg Tiedemann, Uppsala University