Morphological analysis methods for non-dictionary words to improve the accuracy of morphological processing of texts in the russian language

Abstract

The increase in the volume of processed information leads to the necessity of automation of its analysis and development of appropriate software tools. Development of new spheres of people’s activity and misprints lead to the necessity of being able to analyze non-dictionary words. The article describes a comparative analysis of existing methods of morphological processing of unknown words. Based on the results of the analysis, the authors proposed their own algorithm, which solves the existing problems and considers the peculiarities of word formation of the Russian language, thus improving the quality of morphological analysis of individual words and the accuracy of homonymy resolution.