Search

This study lies in the field of Computational Linguistics and is a part of a broader research project that aims to develop an analysis system of Modern Greek applied in translation. It is widely known that during the translation process, the structure both of the source and the target language should be taken into consideration. However, natural languages present various semantic and syntactic characteristics such as polysemy, multiword expressions, differences in meaning of words from singular to plural etc. which arise problems in translation and mostly in automatic/machine translation. In this paper, we focus on the polysemy of Greek simple nouns. In particular, we will study how semantic features that are added to the lemmas of electronic morphological dictionary of Modern Greek (DelasGR) could improve the results of the translation process. We will present the methodology and criteria according to which we add semantic features to the nouns and provide some representative examples of translations taken by the machine translation system Systran., Syn-Thèses, No 5 (2012)