Development of an unified meta language of the turkic languages morphology

Abstract

Currently, due to the sharp increase of information amount in natural languages on the Internet and social networks, research and development in the field of computational linguistics is becoming extremely relevant. As is known, computational linguistics is a new scientific field and part of computer science.Computational linguistics includes the Natural Language Proccesing (NLP). Creating a unified metalanguage for Turkic languages (UniTurk) is an important task for processing Turkic languages. An unified metalanguage system will allow to unify tags, facilitate their understanding and use common software, as well as conduct various studies on linguistic-statistical comparative analysis among the Turkic languages.The article presents some of the results obtained on an international project to create a multilingual ontology and unified metalanguage of the Turkic languages morphology. Using ontological models, the morphological rules of the Turkic (Kazakh, Kyrgyz, Tatar, Turkish, and Uzbek) languages are formalized. The result of these works can be used in the NLP applications, for example, for corpus tagging, in knowledge extraction systems, information retrieval systems, machine translation, etc.