Μέθοδοι εντοπισμού κοινωνιών σε δίκτυα
View/ Open
Keywords
Συσταδοποίηση δικτύου ; Συσταδοποίηση γράφου ; Τεχνικές συσταδοποίησης ; Τμηματοποίηση δικτύουAbstract
Community detection in networks is one of the most popular topics of modern network sci-ence. One of the most basic features of networks, which are represented by graphs, is their com-munity structure. The underlying of community structure in a network, or clustering, meaning the organization of vertices in clusters, with many edges joining vertices of the same cluster and com-paratively less edges joining vertices of different clusters, is a scientific subject which is thor-oughly studied and hasn't stop evolving until today. Network community detection is of great im-portance in sociology, biology or computer science where networks are represented as graphs. The problem of community detection in networks or graphs, is a modern scientific subject, studied from various scientific aspects. In this thesis, we will attempt an in-depth analysis of the subject, from the definition of its main elements for which there is no universal protocol, to the exposition of many methods for community detection which have been proposed throughout the years. In the end we will compare algorithms coming out of different methods we analyzed, using benchmark graphs.