Georgian Language Based Document Classification Method Development

Author: manana khachidze
Co-authors: Magda Tsintsadze Maia Archuadze George Dzamashvili Mariam Vardanidze
Keywords: text processing, text classification
Annotation:

The problem of Information Retrieval is complex and the stage of classification is one of its important initial process. The stemming and lemmatization algorithms appropriate to Georgian language is developed. Particularly the word root defining algorithms for Georgian language were produced. Appropriate software was developed and tested. For the process of testing the collection of Georgian language based text was developed presented in form of database. The text processing modern algorithms were adopted for Georgian text initial compilation and the new algorithm was proposed for consideration. The developed software allowed construction of the appropriate knowledgebase. The part of initial text collection was used for knowledgebase formation.