Serbian Wordnet for biomedical sciences
Sanja Antonic, Cvetana Krstev · 2008
Abstract: Wordnet is a lexical semantic network having as nodes the sets of synonymous word (synsets) which are linked by various semantic relations. The first Wordnet built was the so-called Princeton Wordnet consisting today of approximately 140,000 synsets. Due to its remarkable size and successful inclusion in various computer-based applications it is considered as a de facto standard. In this paper we present the new electronic information resource Serbian Wordnet (SWN) and its initial development. Serbian Wordnet is based on the same principles as Princeton Wordnet (PWN), which means that it has not only inherited its basic structure but also the additional information associated to PWN. The development of Serbian Wordnet started with Balkanet project which was funded by European Commission from (2001-2004). The main goal of the project was development of wordnets for the Balkan languages: Bulgarian, Greek, Romanian, Serbian, Turkish and Czech and their alignment with PWN. The development of Serbian Wordnet continues after Balkanet with its enhancement in various domains. We describe how various traditional and Internet resources are being used for the enhancement of SWN in domain of biomedicine and how we see the possible practical usage of SWN in information retrieval this domain.