Using Transformers and Recurrent Neural Networks for Disambiguating Homonyms in the Georgian Language

Author: Davit Melikidze
Keywords: Transformers, Recurrent Neural Network, Homonym disambiguation
Annotation:

The task of accurately disambiguating homonyms is crucial in natural language processing. Georgian, an agglutinative language belonging to the Kartvelian language family, presents unique challenges in this context. In agglutinative languages like Georgian, words are typically formed by stringing together relatively unchanged morphemes. This structural feature makes the task of disambiguating homonyms much more challenging compared to Indo-European flexive languages. In this talk, we aim to highlight the specific problems concerning homonym disambiguation in the Georgian language and our approach to solving them. We will discuss the unique difficulties presented by the Georgian language and introduce the innovative methods we have developed to address these challenges effectively.