Estonian multiword expressions in computational linguistics Cover Image

Eesti keele püsiühendid arvutilingvistikas: miks ja kuidas
Estonian multiword expressions in computational linguistics

Author(s): Heiki-Jaan Kaalep, Kadri Muischnek
Subject(s): Language and Literature Studies
Published by: Eesti Rakenduslingvistika Ühing (ERÜ)
Keywords: computational linguistics; multiword expressions; multiword expression extraction; lexicon of multi-word expressions; multi-word expression annotation; Estonian

Summary/Abstract: Multiword expressions are known to pose problems for natural languge analysis. By multiword expressions we mean combinations of two or more word(form)s that are habitually used together to express a certain meaning; the term covers both idiomatic and collocational word combinations. This article concentrates on three main tasks in multiword expression processing: extraction, lexicon compilation and annotation. The standard methods for solving these tasks are analysed from the viewpoint of automatic analysis of Estonian, a language with a rich and complicated morphological structure and a free word (or constituent) order.

  • Issue Year: 2009
  • Issue No: 5
  • Page Range: 157-172
  • Page Count: 16
  • Language: Estonian