LINGUISTIC MODELS OF RELATIONAL CONSTRUCTIONS IN THE PARALLEL CORPUS OF THE UZBEK LANGUAGE

This article examines linguistic models of relational constructions in parallel corpora of the Uzbek language, focusing on how syntactic, semantic, and pragmatic relationships are encoded and transferred across language pairs. Parallel corpora serve as valuable resources for analyzing contrastive linguistics, translation patterns, and language-specific encoding of relational meanings. This study investigates various types of relational constructions in Uzbek—including possessive, locative, comparative, and logical relations—and their representation in parallel texts with languages such as English, Russian, and other Turkic languages. The analysis reveals systematic patterns in how Uzbek encodes relational meanings through agglutinative morphology, postpositions, and word order, contrasting with the strategies employed in Indo-European languages. The findings contribute to computational linguistics, translation studies, and the development of more accurate machine translation systems for Uzbek. Methodologically, the research employs corpus-based contrastive analysis, examining alignment patterns and structural divergences across parallel texts. The results demonstrate that while Uzbek relies heavily on case marking and derivational suffixes for expressing relations, parallel languages may employ prepositions, word order, or lexical means, presenting challenges for automatic alignment and translation quality assessment.