We are very pleased to announce that Belgium has officially joined CLARIN as a member!
The Belgian CLARIN consortium CLARIN-BE is led by the Instituut voor de Nederlandse Taal (INT) which is already a CLARIN-B centre: CLARIN technical centre for Flanders. Vincent Vandeghinste is the national coordinator for CLARIN-BE.
Other institutions that currently comprise the National Consortium of Belgium:
- University of Ghent: LT3 ( ), imec-IDLab (speech)
- University of Antwerp: CLIPS (NLP)
- University of Leuven: CCL (NLP), LIIR (NLP), ESAT-PSI (speech).
The focus of CLARIN-BE will mainly be on the inclusion of tools and resources which are currently not present in the CLARIN infrastructure. Work in CLARIN-BE is currently funded through the Flemish CLARIAH-VL Research Infrastructure project, which is a cooperation with the DARIAH Research Infrastructure.
In particular CLARIN-BE will focus on the following:
- Digital Text Analysis Dashboard and Pipeline
- Benchmarking existing NLP tools and new models including (re)training tools on existing resources, such as creating state-of-the art methods for language modelling of historical Dutch or specialized text corpora (e.g. medical text, legal text, etc)
- Benchmarking of NLP tools for natural language understanding
- Integration of NLP tools in Digital Text Analysis Dashboard and Pipeline (DTADP)
- Tools for linguistic preprocessing for Dutch, English, French and German
- Tools providing automatic annotation for tokenisation, lemmatisation, part-of-speech tagging, syntax parsing, chunking and named entity recognition
- Tools for natural language understanding (sentiment analysis, emotion detection, document similarity clustering, topic modelling, stylometry)
- Tools for parallel data (sentence and word alignment).
The CLARIN-BE consortium will introduce itself to the Belgian SSH researcher by organising user involvement sessions at Belgian humanities faculties, such as CLARIN master classes at a conference in digital humanities or social sciences, CLARIN courses at summer schools, or a more specific CLARIN training event. Furthermore, sessions about using the CLARIN infrastructure will be included in several university courses. CLARIN-BE also envisages close cooperation with CLARIAH-NL and will continue to contribute to K-Dutch.