Infrastructure Support
CJVT’s infrastructure offers researchers various services: language resources and tools for Slovene, online platforms (crowdsourcing, gamification), and website hosting for research projects and programmes in linguistics.
LLM4DH project
LLM4DH is a project that combines the latest artificial intelligence technologies to explore the digital humanities. Modern approaches are enabling new insights in linguistics, history, law and folkloristics.The project contributes to the development of interdisciplinary research and contributes new insights for understanding society.
Data completion and gamification of dictionary resources (PODVIG)
The PODVIG project is working on upgrading the data in the Thesaurus of Modern Slovene, the Collocations Dictionary and the Dictionary for Speakers of Slovene as a Second and Foreign Language. In addition, the project will develop an online language game with data from these dictionaries.
Adaptive Natural Language Processing with Large Language Models (PoVeJMo)
The key objective of the Adaptive Natural Language Processing with Large Language Models (PoVeJMo) programme is the development of large-scale language models which have an impact on almost the entire field of artificial intelligence and machine learning.
University licence for Sketch Engine
Center for language resources and technologies allows access to the corpus tool Sketch Engine for all representatives of University of Ljubljana.
PROP – Digitally supported development of writing skills
The aim of the project Empirical foundations for digitally-supported development of writing skills is to support teachers who correct and grade student writing. The solution lies in the digital support of teachers’ work and a digitally supported model of providing feedback to teachers and students.
SPOT – Treebank-Driven Approach to the Study of Spoken Slovenian
The SPOT project focuses on describing the syntactic features of Slovene speech using a new method based on syntactically parsed collections of written and spoken texts (so-called treebanks).
ONLINE NOTES
The project Tolmač (Eng. Interpreter) is focused on developing of a system for automatically translating lectures from Slovene to other languages. Automatic subtitles will help people will hearing loss, and lecture excerpts and recordings will be accessible at a dedicated website.
SLOKIT – Corpus informer and text analyzer
The main purpose of the project is to upgrade the research infrastructure portal CLARIN.SI with services that will bring the content of the portal, especially the corpora, closer to a wider range of users.
Čas za slovenščino
Based on the Čas za slovenščino 1 textbook set, the project has developed a comprehensive interactive learning material for the initial learning of Slovene, which provides learners with a high-quality and enjoyable Slovenian learning experience.
KOST corpus
The Corpus of Slovene as a Foreign Language KOST is a digital collection of texts written by adult speakers for whom Slovene is not their first language.
Collocation dataset
The project produced a training set of 713,310 collocation candidates from the Gigafida 2.0 reference corpus, which were labelled according to their collocational relevance.
DSDE – Development of Slovene in a Digital Environment
The project goal is to meet the needs for computational tools and services in the field of language technologies for Slovene.
SoKol – Synonyms and Collocations 2.0
Upgrading fundamental dictionary resources and databases of CJVT UL, where the Slovene Thesaurus and Slovene Collocations Dictionary will be upgraded into version 2.0
KOLOS – Collocations in Slovene
Basic research into semantic and temporal aspects of collocations, as well as statistics for measuring them.
The Jezikovna Slovenija Site
A website about the language policy of Slovenia, where one can find counsel for questions related to language policy.
KaUč – High-quality Slovene Textbooks
The project is developing quality indicators for textbooks. These will be of practical use in the textbook validation and evaluation process. Tools are also being developed to assist in the definition of the indicators.
NEW GRAMMAR
New grammar of contemporary standard Slovene: sources and methods.
KOMASS
Hungarian-Slovene dictionary concept: from language resource to users.
EMBEDDIA – Cross-Lingual Embeddings for Less-Represented Languages in European News Media
The EMBEDDIA project aims to bring high-end language technologies to under-represented languages in the EU by using cross-lingual embeddings coupled with deep neural networks.
European survey on dictionary use
Identifying user expectations about monolingual dictionaries, their experience in using them, and offered suggestions for improvements.
Corpora Gigafida, Kres,
ccGigafida and ccKres upgrade
New corpora versions: balanced texts, improved accuracy, new user interface …
Thesaurus of Modern Slovene promotion
Informing the public about the new thesaurus, synonymy, and the possibilities of public involvement.
FRENK
Resources, methods, and tools for the understanding, identification, and classification of various forms of socially unacceptable discourse in the information society.
STARK – Statistical analysis of dependency-parsed corpora
STARK is a tool for statistical analysis of dependency-parsed corpora. It returns a frequency lists for dependency trees from dependency-parsed corpora.
Historic versions of the Gigafida Slovene reference corpus
We enabled access to the previous versions of the Gigafida corpus through the online concordancers noSketch Engine and KonText. More specifically, the corpora FidaPLUS, Gigafida 1.0 and Gigafida 1.1 are accessible.
Sloleks lexicon accentuation
The Slovene morphological Lexicon Sloleks was updated with automatic (and partially manually-annotated) accents. Additionally, a new interface which enables crowdsourcing of accent data was developed.
Keywords and n-grams from a textbook corpus
In this project, a corpus of textbooks for primary and secondary school was built. A list of words, n-grams and keywords was extracted.
LIST – efficient Slovene corpus analysis tool
A clear and understandable user interface for the corpusStatistics tool (renamed to LIST) was developed. It enables its users to easily access language statistics in Slovene and other corpora.
TermFrame
TermFrame is a three-year research project (2018-2021) that addresses terminology and knowledge at the intersection of languages, cognition and computer science. The project is funded by the Slovenian Public Agency for Research.
Trojina Institute
Center of language resources and technologies took over the infrastructure maintenance of the Trojina Institute.