Cluster analysis of linguistic profiles of hidden communities

Abstract

The aim of the study is to present clusters of profiles of hidden communities based on linguistic parameters. The article analyzes the structure and relationships between the attributes of clusters of community profiles. The scientific novelty of the study lies in the fact that the combination of methods of hierarchical cluster analysis of hidden network communities and analysis of variance will reveal the uniformity/heterogeneity of the author's texts created at the grammatical and lexical levels. Using the Ward method, three clusters of linguistic profiles were identified, each of which was given a formal Silhouette Score. A meaningful assessment of the profiles is presented in the form of appropriate linguistic comments. As a result of the study, it was found that online publications are characterized by variation at the level of syntax, but not at the level of morphology. The proposed community clustering approach can be used to identify potentially dangerous online subcultures and opinion leaders in the online space. As a result of the implementation of this approach, linguistic profiles of communities are complemented by digital sociodemographic information.