This research demonstrates a method for classifying Telegram channels based on digital profiles, highlighting their roles in information interaction networks.
The purpose of the study: development of a method for determining the digital profile of Telegram channels in information interaction networks and a procedure for classifying channels based on the allocated digital profile. Method: includes the following stages: graph of interacting objects construction, based on data imported from the Telegram network, digital profiles determination for vertices based on their attribute data and graph properties, clustering of vertices based on the selected profiles, centers of the obtained clusters and the original Telegram channels classification, computational experiments and analysis of the results. Results: the article introduces the digital profile definition for a Telegram channel, presented as one of the vertices of a graph of interacting objects. The digital profile is defined through a normalized 5-dimensional feature vector based on the attribute data of the vertex and the graph properties. The selected characteristics reflect the properties of Telegram channels in the constructed graph and the metadata obtained during import from the network. The authors then describe an algorithm for clustering the obtained profiles using configurable parameters. The centers of the selected clusters are classified according to 4 types proposed by the authors, characterizing the roles of vertices in the graph of interacting objects. Due to this, all vertices of the graph are classified – the original Telegram channels of the analyzed network. The proposed approach provides valuable information about the roles of Telegram channels in information interaction networks. Scientific novelty: a new approach to the analysis of Telegram channels is developed: a method for creating a digital profile of a Telegram channel in the form of a 5-dimensional feature vector, allowing the analysis and classification of channels. The approach also proposes a procedure for classifying such digital profiles based on computational methods, which allows to identify the main types of Telegram channels of the downloaded subnet according to a given classification.
No takes yet. Share an insight, caveat, or question.
Popov et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: