Hierarchie komunit v komplexních sítích

Abstract

This bachelor’s thesis investigates the study, implementation, and evaluation of selected methods for hierarchical community detection in complex networks. The aim is to compare their ability to uncover nested structures in both synthetic and real-world data. The theoretical part summarizes existing approaches, while the practical part focuses on methods based on modularity optimization, graph-defined distances, vector space, and probabilistic models. Experiments were conducted on synthetically generated hierarchical networks and real networks, and the results were evaluated using internal and external metrics. The thesis also includes software enabling the generation of hierarchical networks, detection of hierarchical communities, evaluation of the resulting hierarchy, and visualization of the outcomes.