摘要：The recently created RuWordNet thesaurus reflects hierarchical, primarily synonymous, relations in the vocabulary so that it is built from a set of synsets. In this paper, we applied the method of a comparative analysis: we compared the data of classical dictionaries of Russian language synonyms, RuWordNet thesaurus and the results of a students' poll. The students represent two areas of specialization: linguistic and nonliguistic. As a result of the psychosemantic experiment, we specified the criteria for the distribution of lexical units among synsets. We referred to them cognitive, semantic, stylistic and quantitative criteria. The quantitative method made it possible to determine the fixation frequency of synonyms for the stimulus word in the respondents' minds. The results of the experiment showed that the most adequate principle, implemented in verbal and cogitative activity by the identifying of the synonymic chains, is a pragmatic one. It reveals itself in the possibility of the interchange of similar in meaning words in different contexts. The research results contain recommendations on the systematization of linguistic information in the RuWordNet thesaurus, aimed at improving the principles of synsets' building. The performed research is aimed at verifying the reflected in RuWordNet data from the point of view of their correspondence to the real idea of native speakers of the synonymic, hyponymic and other system-based relationship in the vocabulary.
其他摘要：Объектом изучения стал недавно созданный тезаурус RuWordNet, который отражает иерар- хические отношения в лексике (прежде всего синонимические) и состоит из совокупности синсетов. Иссле- дование нацелено на верификацию данных, зафиксированных в RuWordNet,
