The article primarily focuses on the DTD/schema which lies at the core of the web portal. It enables the unifying of heterogeneous terminological data and provides a very basic initial markup for new terminology resources: TBX database available on the web.
B.03 Paper at an international scientific conference
COBISS.SI-ID: 37644898The existing Slovene language resources should be supplemented with new ones, among them also terminological resources. A significant number of terminological resources exist, but they are methodologically heterogeneous and often unavailable to the general public. The main goal of the Slovene Terminology Web Portal project is to unify terminology resources in a single termbase.
B.05 Guest lecturer at an institute/university
COBISS.SI-ID: 36800354The paper presents automatic term extraction from mono- and multilingual corpora. The introductory part of the paper outlines the main methods employed in term extraction utilizing both statistical and linguistic corpus data. Since the introduction of the first freely available morphosyntactic annotation tool for Slovene, methods in term extraction have been primarily based on morphosyntactic patterns which help us extract potentially relevant terms from the corpus. These candidate phrases are subsequently ranked according to a statistical measure of termhood.
B.03 Paper at an international scientific conference
COBISS.SI-ID: 28535597Young researcher PhD supervisor
D.09 Tutoring for postgraduate students
COBISS.SI-ID: 40243810Young researcher PhD supervisor
D.09 Tutoring for postgraduate students
COBISS.SI-ID: 40174690