Об особенностях построения и анализа графов взаимодействующих объектов в сети Telegram-каналов

Вопросы кибербезопасности. 2023. № 1(53). С. 75–81.

Chepovskiy A.

The purpose of the study: search for a technique for constructing and analyzing a graph of interacting objects in the network of Telegram channels, including the calculation of psycholinguistic characteristics of texts. This technique makes it possible to classify groups of channels and evaluate their informational impact on users.

Method: (U, M, R)-model is used to build a weighted graph during data import. Next, on the resulting graph, the Galaxies method is applied to reveal implicit intersecting communities. Psycholinguistic factors are calculated on the imported combined texts of communities to assess the channels thematic focus.

Results: the article presents a methodology for working with a network of Telegram channels in order to identify
groups of channels that carry out information impact on users. A full cycle of actions is presented, starting from
data import, using a model for constructing a graph of interacting objects for such networks, and ending with the
calculation of psycholinguistic characteristics of texts for groups of channels. At the same time, the issue of the
most effective selection of implicit communities in networks of Telegram channels is highlighted. An example of
a network and a constructed weighted graph with markers calculated on texts, which are the most indicative for
identifying the channels focus, is presented. The presented approach, by highlighting significant differences in the
corresponding markers, makes it possible to identify channels that most actively carry out informational impact on
users.

The scientific novelty: The combination of an algorithmic approach and the use of psycholinguistic research represent the scientific novelty of this method. The results obtained make it possible, using the methods of computational linguistics in combination with the communities reveal methods, to evaluate different participants in such networks.

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