Выделение неявных пересекающихся сообществ на графе взаимодействия Telegram-каналов с помощью «метода Галактик»

Труды Института системного анализа Российской академии наук. 2022. Т. 72. № 4. С. 39–50.

Popov V., Chepovskiy A.

In this paper, the authors present the “Galaxies method” to reveal implicit communities on the
graph of interacting objects obtained by importing a network of channels from the Telegram messenger. This
method is based on successive identification of overlapping communities on the initial weighted graph, further
construction of a new graph, in which the vertices are the communities revealed at the first step, called by
the authors “metavertices”. The weighted edges of the new graph between the “metavertices” are built based
on the weights between each pair of vertices in the original graph. Further, non-overlapping communities are
identified on the new graph. The result is a partition of the original graph into overlapping “meta-communities”.
To assess the quality of partitioning using the presented method, the authors carried out a psycholinguistic
analysis of the obtained metacommunities, identified patterns depending on the thematic orientation of channels
within the metacommunity. As a result of processing and analysis of the obtained metacommunities, the quality
of the partition was confirmed. The combination of the algorithmic method of revealing communities and
psycholinguistic analysis has a practical application for the analysis of information impact in social networks.

Keywords: анализ социальных сетей, analysis of social networks, выделение сообществ, community detection, telegram, импорт данных из социальных сетей, data import from social networks, model of information impact, graph of interacting objects, psycholinguistic analysis of texts, Telegram, модель информационного воздействия, граф взаимодействующих объектов, психолингвистический анализ текстов