A Hierarchically Integrated Annotation Model for Literary Texts: Theoretical Foundations and Empirical Validation on Uzbek Prose

Digital literary study relies on linguistic annotation, but standard models fail to integrate textual layers. This paper introduces the Hierarchically Integrated Annotation Model for literary texts. Synthesizing seven analytical levels on a token reference system, HIAM unifies major international systems with Uzbek corpus-linguistic standards. Tested on a 1000-token 20th-century Uzbek prose sample, the model achieved 90.9% accuracy and strong consensus (\(\kappa = 0.85\)), proving its viability for computational philology in agglutinative languages.

Перевод пока недоступен