FEATURES AND DIFFERENCES OF USING GRAPH THEORY FOR ANALYSIS AND VISUALIZATION OF CONNECTIONS IN SOCIAL NETWORKS

Abstract

This article explores the potential for adapting classical graph theory to the analysis of social networks through key criteria such as node types, edge types and weights, graph directionality, temporal variability, centrality metrics, community structures, analytical algorithms, big data processing, visualization, prediction capabilities, and the role of content. The aim of the research is to enhance the effectiveness of analyzing the structure and dynamics of social networks by applying graph theory methods and visualization tools to identify interaction patterns between users.The paper provides a comparative analysis of how graph theory is applied to study the structure and user interactions in four of the most popular social networks: Facebook, Instagram, TikTok, and LinkedIn. It discusses differences in graph structure (nodes, edges, edge weights, directionality), connection types and their evolution over time, the nature of user interactions, analytical approaches (methods and forecasting), big data handling, and visualization techniques.The key analytical methods used to examine connections within each of the selected networks are outlined, and visualizations of graphs representing these connections are presented. The study demonstrates that the defined graph- theoretical features for analyzing the structure of leading social networks can significantly improve the use of these tools in the areas of: ensuring cybersecurity, supporting political interest analysis through public opinion monitoring, enhancing customer relationship management, and optimizing social media algorithms.Data sources for this type of analysis may include: direct platform data; data from electronic communication systems (e.g., email, chats, forums); internal organizational records; results of surveys conducted to map social ties and user interactions; and publicly available datasets.The scientific novelty of the study lies in the development of a methodology for selecting appropriate types of graph models and visualization strategies for analyzing the structure and dynamics of multilevel social connections with maximum efficiency. The results obtained may be applied by government institutions, business entities, and civil society organizations to gain deeper insights into digital social interactions – not only at the level of basic connections (“friendship”, “subscription”), but also in more complex informational and behavioral contexts.