PRINCIPLES OF USING LINGUISTIC RESOURCES IN THE SENTIMENT ANALYSIS PROCESS OF UZBEK TEXTS

This article deeply explores the principles of utilizing linguistic resources in the sentiment analysis process of Uzbek texts. Considering the agglutinative nature of the Uzbek language, vowel harmony, and cultural context, the role of lexical databases, corpora, and morphological analyzers is analyzed. The article examines the stages of data preparation, feature extraction, classification, and aspect-based sentiment analysis (ABSA). It also discusses challenges in low-resource languages and future prospects, including hybrid approaches and the development of language models. These principles, grounded in empirical and theoretical methods of linguistics, serve to enhance the accuracy and applicability of the analysis.