Analysis of approaches to the automatic extraction of contextual synonyms from texts in the russian language

Abstract

The article addresses the problems of determining contextual synonyms and provides the results of the analysis of approaches to their automatic extraction from texts in the Russian language. Several approaches are proposed. These approaches are based on the use of linguistic algorithms to determine similar semantic-syntactic structures in texts and their combination with machine learning methods. The results from the application of the algorithms were analysed. The algorithms were based on morphological, syntactic, and semantic analysis of the text, filtering the results by keywords, and using various clustering tools. In the conclusion, the applicability of the implemented approaches was discussed, and the directions for the development of a combination of these approaches were determined.