RaynorJim / social-networks-analysis-wan-bms

Analysing social networks, both real and online using actual data

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Bridging Online and Offline Social Networks: Multiplex Analysis


We study a unique dataset that looks at student friendships in real life and on Facebook. We consider two types of friendships: strong and weak, which are defined differently for real and Facebook friendships. For real friends, 5 contacts per month is considered a strong tie, while less than that but more than 5 contacts in six months is considered a weak tie. On Facebook where communication is more frequent to friends are considered strongly connected if they communicate at least 5 times a weak. Any less than that, but still more than 5 times a month and the Facebook friendship is considered as weak.

Based on this four types of network ties we define a multiplex network with 4 layers: strong offline, weak offline, strong online, weak online.

alt text

A visualization of the 4 layers using MuxViz.

The resulting multiplex network is analysed using simple graph metrics such as reciprocity and transitivity that are expanded to mutliplex networks. For more details please see the supporting paper.

alt text

Three cycles metric calculated on 6 layers of the multiplex graph (4 original, 2 aggregated).

About

Analysing social networks, both real and online using actual data


Languages

Language:Python 88.2%Language:TeX 11.8%