Social Network Analysis of the BitTorrent Network

State: completed by Romana Pernischova

Social Network Analysis (SNA) traditionally investigates individuals and their social relations. Individuals and realtions are typically modelled as a graph consisting of nodes (individuals) and edges (relation, e.g., friendship). Modelling a social network as a graph enables the calculation and interpretation of several graph metrics, such as Centrality Metrics, Symmetry Measures, and Entropy Measures [3]. Libraries for the calculation of these measures are available for platforms [5] and can, therefore, be easily applied to an accordingly prepared data set. However, the interpretation of these measures strongly depends on the mapping chosen to model an observation as a graph. SNA methods are not limited to social networks [2], in fact, these can be applied to any network that can be modelled by nodes and edges. The Orgnet definition of SNA: “Social network analysis [SNA] is the mapping and measuring of relationships and flows between people, groups, organizations, computers, URLs, and other connected information/knowledge entities.” [4] gives hints on the applicability of SNA to areas beyond Social Networks.

This seminar report will investigate how SNA can be applied to P2P file sharing systems. BitTorrent (BT) shall serve as example application since it is the most popular file P2P file sharing application. Several possibilities of modelling the BT network as a graph exist. The main goal of this report is to identify these possibilities and find interpretations for the common SNA measures. Depending on the abstraction the interpretation of the SNA measures will vary. Therefore, the report requires a section which details SNA measures and their generic interpretation which will lated be applied to the concrete model abstracted from the BT network.


“BitTorrent.org”. http://www.bittorrent.org/index.html

Burger, Valentin, et al. "Social Network Analysis in the Enterprise: Challenges and Opportunities." Socioinformatics-The Social Impact of Interactions between Humans and IT. Springer International Publishing, 2014. 95-120.

Hoßfeld, Tobias, et al. "On the computation of entropy production in stationary social networks." Social Network Analysis and Mining 4.1 (2014): 1-19.

Orgnet, “Social Network Analysis, A Brief Introduction”. http://www.orgnet.com/sna.html

The igraph core team, “igraph”. http://igraph.org/

Final Report

30% Design, 30% Implementation, 20% Analysis, 20% Documentation
Good programming skills (Java or similar) / datamining experience, background in statistics

Supervisors: Andri Lareida

back to the main page