Blog post written by Tomaž Erjavec, Jakob Lenardič, and Nikola Ljubešić, edited by Darja Fišer
CLARIN.SI joined CLARIN in 2015. The seat of CLARIN.SI is the Jožef Stefan Institute, the main Slovenian research organization for applied research in natural sciences and technology, where three units are involved in overseeing its operation: the Department of Knowledge Technologies, the Laboratory for Artificial Intelligence, and the Centre for Networking Infrastructure. CLARIN.SI is organised as a consortium, bringing together partners from all the main organisations that produce or use language resources in Slovenia, in particular the four Slovenian universities (University of Ljubljana, University of Maribor, University of Nova Gorica and University of Primorska), three research institutes (Scientific and Research Centre of the Slovenian Academy of Sciences and Arts, Jožef Stefan Institute, Institute of Contemporary History), three societies (Slovenian Language Technologies Society, Trojina Institute for Applied Slovene Studies, Domestic Research Society), and two HLT companies (Alpineon and Amebis). The national coordinator of CLARIN.SI is Tomaž Erjavec.
CLARIN.SI has very good relations with similar research infrastructures in Slovenia, in particular DARIAH-SI, and ADP, the Slovenian CESSDA node, which are realised through joint work on specific projects, such as the recent ParlaFormat workshop, and partnership in the recently started RDA-Slovenia project.
CLARIN.SI is a B-certified centre which offers a LINDAT/D-Space repository that currently contains around 110 language resources for Slovenian as well as for other languages, especially Croatian and Serbian. The repository offers a wide range of large corpora for linguistic research on Slovenian, as well as parallel and manually annotated corpora and lexica for training language tools. Most of the corpora in the repository can also be accessed via two concordancers, KonText and noSketch Engine, both of which are integrated with the repository and serve as versatile online environments for searching and efficiently analysing large and richly annotated corpora. In addition to resources, the centre offers tools for text processing as well, either as open source on GitHub, or as on-line services, such as ReLDIanno, an on-line tool and web service for, currently, annotating texts in Slovenian, Croatian, and Serbian.
The consortium regularly supports data curation projects, mostly in terms of annotation campaigns, or to prepare existing digital data for inclusion into the repository. A good example of in-kind support is the Kontext.io semantic lexicon of Slovene, Croatian and Serbian, where CLARIN.SI prepared the union of its public corpora, in order to train the word embeddings that are the basis of the lexicon, as well as providing examples of use on the portal. In 2018, support for ad-hoc projects was supplemented by a project call to the consortium partners, through which seven projects were selected for financing. All the projects produced openly available resources or tools for Slovenian, and the call has been repeated in 2019, with six projects accepted for funding.
A major priority of the Slovenian consortium are its outreach activities, many of which have an international scope. The Slovenian Language Technologies Society has been organising biennial conferences on Language Technologies with on-line reviewed proceedings for over 20 years. In 2016 the scope of the conference was extended to Digital Humanities, and CLARIN.SI became one of the organisers and supporters of the conference. The 11th edition of the Language Technologies & Digital Humanities conference, which took place in September 2018 in Ljubljana, heard presentations of 47 papers (21 papers in Slovene and 26 in English), including 2 talks by invited lecturers. The Society also organises, for almost 15 years, regular “JOTA” lectures on language technologies; since 2017 CLARIN.SI has supported the recording of these lectures, which are available, together with video-synchronised slides on the VideoLectures portal. CLARIN.SI supports other events that take place in Slovenia and are related to the mission of CLARIN, e.g. in 2018 CLARIN.SI this was the XVIII EURALEX International Congress, and in 2019 the 22nd International Conference “Text, Speech and Dialogue”.
Recently, CLARIN.SI has established the Knowledge Centre for South Slavic languages (CLASSLA). CLASSLA offers expertise on language resources and technologies with its basic activities being (1) giving researchers, students, citizen scientists and other interested parties information on the available resources and technologies via its documentation, (2) supporting them in producing, modifying or publishing resources and technologies via its helpdesk and (3) organizing training activities. The K-Centre also offers a FAQ for Slovene, Croatian and Serbian and documentation on how to use ReLDIanno CLARIN.SI web services.
The vibrant CLARIN.SI community gathered at the 11th Slovenian Language Technologies and Digital Humanities Conference in 2018
Click here to read more about Tour de CLARIN