State-of-the-Art NER Models for the Uzbek Language

Named Entity Recognition (NER) is one of the fundamental tasks in Natural Language Processing and plays an important role in many applications such as information extraction, machine translation, and question answering systems. In recent years, machine learning and deep learning approaches have significantly improved the performance of NER systems. However, developing accurate NER models for low-resource languages such as Uzbek remains a challenging task due to the limited availability of annotated corpora and linguistic resources. This paper reviews state-of-the-art NER models used in machine learning for the Uzbek language, including traditional statistical methods and modern neural network architectures. In particular, models based on Conditional Random Fields, Bidirectional LSTM, and transformer-based architectures such as BERT and XLM-RoBERTa are analyzed. The study discusses their effectiveness, advantages, and limitations in the context of Uzbek language processing. The findings highlight that transformer-based multilingual models demonstrate the best performance for Uzbek NER tasks and provide promising directions for future research.