title: Concept based resources and processing tools for the Estonian language
reg no: ETF5534
project type: Estonian Science Foundation research grant
subject: 6. Humanities
status: accepted
institution: TU Faculty of Philosophy
head of project: Haldur Õim
duration: 01.01.2003 - 31.12.2006
description: The main goal of the present grant is to formulate a systematic overview of recent trends of development and new tools of concept-based language processing, and their requirements as applied to the Estonian language. The concept-based approach to language processing has undergone a very rapid progress in recent years and as its result a quite new situation in the relationship between language technology and theoretical linguistics has arised. This has caused a need to get an overwiev of these developments and to create a programme for their application in the Estonian language processing. The second goal is to carry out concrete research in areas which in any case form the basis of concept-based language processing: 1. semantic metalanguage (ontology): categories, their relationships; 2. relationship between syntax and semantics (semantics of syntactic constructions); 3. semantics of word classes (e.g. verbs, adjectives); 4. word sense disambiguation programs. These tasks presuppose remarkable development of the following excisting semantic resources of Estonian: 1) semantic database (wordnet): at least to 30 000 entries; 2) semantically disambiguated text corpus - at least to 100 000 words.

project group
no name institution position  
1.Neeme KahuskUniversity of Tarturesearcher 
2.Kaarel KaljurandUniversity of TartuPh.D. student 
3.Heili OravUniversity of Tarturesearcher 
4.Kadri ViderUniversity of Tarturesearcher 
5.Haldur ÕimTU Faculty of Philosophyprofessor