Curent research projects
- Development of a curriculum for medical interpreters (MedInt): The main objective of the international project MedInt is to initiate a process of exchange of ideas and experience among the project partners which will result in a joint curriculum for training healthcare interpreters. The project coordinator is the Institute for Theoretical and Applied Translation Studies, University of Graz (Austria), while the partners are the following: departments of translation at the University of Ljubljana, the University of Germersheim (Germany), the University of Tampere (Finland), KAGes (Clinical Center in Graz), and ZEBRA (therapeutical centre in Graz). The Department of Translation, University of Ljubljana, is represented by Nike K. Pokorn and Lars Felgner. The project’s webpage: http://www.medical-interpreting.eu/
- Slovene terminology web portal: The main objective of the project is to develop a terminology portal which collects terminological data from various sources, provides basic information on the principles of terminological work, and offers various tools for termbase standardization and conversion into an open standard format TBX. The core functionality thus consists of a series of conversion modules that transform different types of existing terminology data from different sources into an XML format with a simple DTD/schema and from there to a unified TBX database. Additional functions of the portal are a term extraction and a term annotation tool.
- Comparative evaluation of the Hungarian and Slovene wordnet in machine translation: The chief motivation of the bilateral project in which the department is collaborating with the Hungarian Academy of Sciences from Budapest is to test the applicability of the recently developed Slovenian and Hungarian wordnets in one of the most active research fileds in NLP: machine translation. We believe that WordNet may significantly improve Machine Translation mainly in the Word Sense Disambiguation segment, but also in cases where no translation equivalent is found for a source word or phrase, as the semantic database may provide a hypernym.
- JOS project. Linguistic Annotation of Slovene: The JOS project is run by the Jožef Stefan Institute and its goal is the development of Slovene annotated corpora and associated resources for Human Language Technologies for the Slovene language. The department participates in the activities related to semantic annotation of the collected corpora. The project’s webpage: http://nl.ijs.si/jos/index-en.html
