Table 4.

Area Under Curve (AUC) scores for link prediction. Comparison with popular baselines and embedding based methods bootstapped using binary operators: (a) Average, (b) Hadamard, (c) Weighted-L1, and (d) Weighted-L2 (See Table 1 for definitions).

Op	Algorithm	Dataset
		Facebook	PPI	arXiv
	Common Neighbors	0.8100	0.7142	0.8153
	Jaccard’s Coefficient	0.8880	0.7018	0.8067
	Adamic-Adar	0.8289	0.7126	0.8315
	Pref. Attachment	0.7137	0.6670	0.6996
	Spectral Clustering	0.5960	0.6588	0.5812
(a)	DeepWalk	0.7238	0.6923	0.7066
	LINE	0.7029	0.6330	0.6516
	node2vec	0.7266	0.7543	0.7221
	Spectral Clustering	0.6192	0.4920	0.5740
(b)	DeepWalk	0.9680	0.7441	0.9340
	LINE	0.9490	0.7249	0.8902
	node2vec	0.9680	0.7719	0.9366
	Spectral Clustering	0.7200	0.6356	0.7099
(c)	DeepWalk	0.9574	0.6026	0.8282
	LINE	0.9483	0.7024	0.8809
	node2vec	0.9602	0.6292	0.8468
	Spectral Clustering	0.7107	0.6026	0.6765
(d)	DeepWalk	0.9584	0.6118	0.8305
	LINE	0.9460	0.7106	0.8862
	node2vec	0.9606	0.6236	0.8477