Development and Realization of Bigram Models for Recognizing Homonyms in the Uzbek Language

Homonyms are words that have the same form but a different meaning. These types of words have always been considered to be of interest for linguistic developments in the course of general linguistics. In this regard, homonyms found in texts are separately studied in Russian and European linguistics in connection with the linguistic corpus. This article analyzes methods and models in linguistic programs and systems based on linguistic software in world linguistics, such as Brill’s method, hidden Markov model, modification of models and their connection with the linguistic corpus, and provides insights into the importance of the national corpus in linguistic processing. In this article, several methods for identifying homonyms in Uzbek language texts were deliberated, and the N-gram method was identified as one of the most reliable methods for the Uzbek language. For this, in work are made 75 models for bigram and trigram ways of connecting words in the Uzbek language.