Advancing Latvian computational lexical resources for natural language understanding and generation

Year 2023 Apr–2026 Mar
Funding Latvian Council of Science
Fundamental and Applied Research Projects
Abstract The goal of the project is to develop Latvian lexical resources and software tools necessary for language understanding and generation. It is planned to expand the Latvian WordNet to a wide coverage resource, to create a semantic system of derivatives and multi-word entities in the largest open explanatory dictionary of the Latvian language, and to create a computable resource in the multilingual Grammatical Framework. The possibilities of using the developed computing resources will be demonstrated with an application example in the automatic generation of medical records. In the linguistic aspect, the project includes fundamental research on the semantics of derivations and the structure and semantics of multi-word expressions, while in the computational linguistic aspect, the understanding and generation of natural language is developed using the lexical resources created during the project.


I. Lokmane, L. Rituma, M. Stade, A. Klints
Paradigmatiskās semantiskās attieksmes „Latviešu valodas leksiskajā tīklā” („Latvian WordNet”)
Linguistica Lettica, 31(1), 126-149, 2023
M. Grasmanis, P. Paikens, L. Pretkalnina, L. Rituma, L. Strankale, A. Znotins, N. Gruzitis
Tē – the experience of building a multifunctional lexical resource
Electronic lexicography in the 21st century (eLex 2023): Invisible Lexicography. Proceedings of the eLex 2023 conference, Lexical Computing CZ s.r.o., 2023