Research Meeting 13043
Workshop "Annotation and Alignment of Parallel Corpora for Linguistic Research"
( Jan 22 – Jan 25, 2013 )
Permalink
Organizers
- Gintare Grigonyte (Universität Zürich, CH & Vytautas Magnus University - Kaunas, LT)
- Ruta Marcinkeviciene (Vytautas Magnus University - Kaunas, LT)
- Andrius Utka (Vytautas Magnus University - Kaunas, LT)
- Martin Volk (Universität Zürich, CH)
Contact
This workshop is centered around the topic of multilingual parallel corpora. It aims at fostering the understanding of opportunities that arise from automatic word alignment for bilingual texts. Because of statistical word alignment methods, it is now possible to build large word-aligned parallel corpora. These constitute a new way of accessing lexical data in context. Alignment search systems like Linguee or Bilingwis help not only in language learning (tekstynas.vdu.lt), but also in translation studies, terminology, and lexicography. They allow for the discovery of previously undocumented translation alternatives and word senses with frequency distributions. When enriched with linguistic information such as named entity classes, syntactic trees or co-reference tags, they open new possibilities for corpus linguistics.
This Workshop is supported by the Sciex NMS-CH programme of the Rector's Conference of the Swiss Universities (CRUS) under project code 11.002.
Keywords
- Machine translation
- Translation studies
- Terminology and lexicography
- Multilingual text analysis
- Contrastive Corpus Linguistics
- Alignment and Search Tools assessment