A PREDICATIVE-CENTERED FORMAL-LOGICAL MODEL FOR AUTOMATIC IDENTIFICATION OF SIMPLE SENTENCES IN THE UZBEK LANGUAGE

Abstract

This article addresses the problem of automatically differentiating simple and complex sentences in Uzbek texts. Unlike traditional approaches, it proposes the number of predicative centers, rather than the number of verb forms, as the principal criterion, and develops a formal model based on the chain "verb form → predicate candidate → independent predicate → predicative center"; this model determines the sentence's structural type (simple, complex, or a construction without a predicative center), and proposes the formula MSG = P ∧ (A ∨ U ∨ K) for the complicated simple sentence. The second part of the article describes a rule-based logical model, in which the sentence's state is expressed as a vector of morphological, predicative, syntactic, and contextual features, and a decision is made on the basis of a rule base with priority coefficients. The model is checked contrastively on Uzbek and English material: in both languages the unity of the predicative center is preserved, but the formal features expressing it depend on the language's structure (morphological markers in Uzbek, analytic means in English). The model forms the theoretical basis of the algorithm and the UzSentence Analyzer tool described in the following article.