Computational Model of the Modern Georgian Language and Search Patterns for an Online Dictionary of Idioms

Abstract

This paper describes the results of projects implemented under the financial support of the Shota Rustaveli National Science Foundation (Nos. DP2016_23, LE/17/1-30/13, AR/320/4-105/11, Y-04-10), particularly the use of finite state technology, specifically lexc and xfst, for the morphological analysis of the Modern Georgian language and the application of a morphological transducer to solve problems of lemmatization and alphabetization noticed in Georgian dictionaries. For instance information on lemmas and the morphological structures of words determined by means of the transducer was used to solve the above-mentioned lexicographic problems in the Online Dictionary of Idioms available at http://idioms.iliauni.edu.ge/. The Online Dictionary of Idioms is monolingual (Georgian) and bidirectional bilingual (Modern Georgian – Modern Greek and vice versa), and is equipped with a friendly environment to find idioms, their meaning and context of use. Its compilation followed the main principles as stated by Gibbon & Van Eynde (2000) regarding the linguistic specification, database management system, phases of lexicographic database construction and access to lexical information.