Developing NLP Tool for Linguistic Analysis of Turkic Languages

Nilufar Abdurakhmonova, A. Ismailov, Davlatyor Mengliev

20222022 IEEE International Multi-Conference on Engineering, Computer and Information Sciences (SIBIRCON)19 citationsDOI

Abstract

Today we see the active development of natural language processing technologies, including the morphological analysis of word forms. In this context, the development of more universal (general) algorithms that we can work with several languages at once is of particular relevance. As a result, algorithms that can reliably process texts of one family will be much more popular than those that can work with only one language. In our work, we presented a flexible tool for working with Turkic languages, in particular, for the morphological analysis of words. As is known, the Turkic-speaking languages belong to agglutinative languages, which, in turn, are formed by the concatenation of morphemes. It follows from this that by means of various combinations of affixes it is possible to construct a huge number of words in each of the languages of this family. Our tool allowed to carry out morphological analysis for almost all Turkic languages.

Topics & Concepts

Agglutinative languageComputer scienceMorphemeConcatenation (mathematics)Natural language processingLanguage familyArtificial intelligenceLinguisticsProcess (computing)Context (archaeology)Relevance (law)Construct (python library)MathematicsHistoryProgramming languagePhilosophyCombinatoricsArchaeologyLawPolitical scienceNatural Language Processing TechniquesLinguistics and Cultural StudiesTopic Modeling