Language Models and Resources
LaMoRe is a scientific research group at the University of Bergen. Its mission is to do research on computational language models and digital language resources. Its scientific basis is in general and formal linguistics, computational and corpus linguistics, psycho- and neurolinguistics, lexicography and Artificial Intelligence. LaMoRe is a national leader for research infrastructure on language data and technologies.
Central themes for the research group are the following:
Linguistic description and formalisms. The LaMoRe research group investigates the grammar and lexicon of languages of various families, including sign languages.
Models of natural language processing by humans and machines. LaMoRe performs experiments and does computational modeling including Artificial Intelligence (AI) models based on deep learning.
Construction and use of digital language resources. Language resources of interest to LaMoRe are text and speech corpora annotated at various linguistic levels (including treebanks), electronic dictionaries and lexicons, thesauri and wordnets, termbanks, taggers and grammars.
Language technology applications. LaMoRe is involved in R&D aimed at applications such as lexicography, machine translation, information search, document classification and text processing tools.
The research group cooperates with a large number of institutions in 34 countries.