Ensemble clustering for graphs: comparisons and applications

Applied Network Science

Table 1 We run each clustering algorithm 100 times on the college football dataset, namely: ECG, Louvain (ML), WalkTrap (WT) and InfoMap (IM)

Results for the college football graph
	ECG	ML	WT	IM
ARI	.889 ±.016	.763±.052	.815	.897
AMI	.900 ±.005	.843±.020	.856	.899
AGRI	.869 ±.005	.815±.019	.837	.872

We compute three different measures with respect to the ground-truth communities (12 conferences). Where significant, we also report the standard deviation. Best results in bold