Simulating systematic bias in attributed social networks and its effect on rankings of minority nodes

Applied Network Science

Table 2 Overview of the empirical datasets

Name	Topic	Nodes	Edges	\(Q_{mod}\)	f	A
brazil (Rocha et al. 2010)	Sexworker	15k	38k	\(-\)0.500	39%	\(-\)1.00
pok (Holme et al. 2004)	Online dating	25k	25k	\(-\)0.349	43%	\(-\)0.84
github (Karimi 2019)	Followers	119k	248k	0.004	5%	\(-\)0.03
dblp (Karimi et al. 2016)	Coauthorship	185k	619k	0.027	21%	\(-\)0.15
aps (Karimi 2019)	Citation	1k	3k	0.346	32%	0.74

We use \(Q_{mod}\)/A to denote the modularity/assortativity of the partition based on minority and majority label. The modularity is not obtained from an optimisation. Both metrics may be interpreted as a proxy for the homophily in the dataset. The fraction of the minority in the dataset is denoted by f. Our datasets cover a wide range of sizes, minority fraction and modularity/assortativity