Models for Effective Categorization and Classification of Texts into Specific Thematic Groups (Using Gender and Criminal Themes as Examples)

Abstract

An analysis of existing automated methods for text classification, used to develop an effective approach for automated text classification by thematic groups in the context of information related to criminal and gender themes, was conducted. Based on the analysis of classification methods, an algorithm for classifying texts by types of crime and gender was developed, information-linguistic and software for the task of distributing texts into thematic groups were developed, and the effectiveness of the developed application was assessed.