engleski

Synonymy between theory and practice: The corpus-based approach to determining

The aim of this research is to determine the degree of synonymy among words based on a corpus search in hrWaC 2.0 (Croatian Web Corpus, Version 2). First, we analyze the lexicographic determination of absolute syn- onymy at the level of the dictionary definition (where the independent meaning of the words is extracted from usage context). The major part of the research focuses on the usage of synonyms based on corpus data. The analysis includes two pairs of synonymous nouns in Croatian: krevet and postelja ‘bed’ and bijes and gnjev ‘rage’. The ten most common multiword units (MWUs) of these syn- onyms were selected as attested in the corpus according to raw frequency. According to the obtained results, we determined how many synonyms are distanced from each other when compared to other words, whereby these words obtain the attributes of partial synonyms instead of absolute synonyms. Despite the fact that the context permits the use of both synonyms, it was confirmed that only one is significantly prevalent in some MWUs. The corpus-based approach imposes a different manner of defining absolute synonyms in dictionaries – apart from describing the basic semantic features of the word, the definition must consider the contextual representation and the collocational perspective of synonymous words. Corpus results can improve synonymous relationships among words in lexicographic descriptions.

Corpus-based Approach to Synonymy, Croatian, Lexicographic Synonymy, Multiword Units

nije evidentirano