Skip to main navigation Skip to search Skip to main content

Natural language processing for the swiss german dialect area

  • University of Geneva

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

12 Scopus citations

Abstract

This paper discusses work on data collection for Swiss German dialects taking into account the continuous nature of the dialect landscape, and proposes to integrate these data into natural language processing models. We present knowledge-based models for machine translation into any Swiss German dialect, for dialect identification, and for multi-dialectal parsing. In a dialect continuum, rules cannot be applied uniformly, but have restricted validity in well-defined geographic areas. Therefore, the rules are parametrized with probability maps extracted from dialectological atlases.

Original languageEnglish
Title of host publicationSemantic Approaches in Natural Language Processing - Proceedings of the Conference on Natural Language Processing 2010, KONVENS
StatePublished - 2010
Event10th Conference on Natural Language Processing, KONVENS 2010 - Saarbrucken, Germany
Duration: Sep 6 2010Sep 8 2010

Publication series

NameSemantic Approaches in Natural Language Processing - Proceedings of the Conference on Natural Language Processing 2010, KONVENS

Conference

Conference10th Conference on Natural Language Processing, KONVENS 2010
Country/TerritoryGermany
CitySaarbrucken
Period09/6/1009/8/10

Fingerprint

Dive into the research topics of 'Natural language processing for the swiss german dialect area'. Together they form a unique fingerprint.

Cite this