Development of Algorithms and Software for Syntactic Analysis of Uzbek Language Texts, and Creation of A Dependency Parser Model for Uzbek Language

This paper addresses the development of algorithms and software for syntactic analysis of Uzbek language texts and the construction of a dependency parser model tailored to the linguistic characteristics of Uzbek. Given the agglutinative nature and relatively free word order of the Uzbek language, traditional rule-based approaches are insufficient for achieving high parsing accuracy. Therefore, this study integrates probabilistic models, graph-based parsing techniques, and neural network architectures to enhance syntactic parsing performance. The proposed model leverages morphological features and contextual embeddings to capture syntactic dependencies effectively. Experimental results demonstrate that the developed system achieves competitive accuracy and robustness, making it suitable for real-world applications such as machine translation, information retrieval, and intelligent dialogue systems.