PubMed

This is the bipartite document–word dataset of PubMed. Left nodes are documents and right nodes are words. Edge weights are multiplicities.

Metadata

Code		`PM`
Internal name		`bag-pubmed`
Name		PubMed
Data source		http://archive.ics.uci.edu/ml/datasets/Bag+of+Words
Availability		Dataset is available for download
Consistency check		Dataset passed all tests
Category		Text network
Node meaning		Document, word
Edge meaning		Occurrence
Network format		Bipartite, undirected
Edge type		Unweighted, multiple edges

Size	n =	8,341,043
Left size	n₁ =	8,200,000
Right size	n₂ =	141,043
Volume	m =	737,869,083
Unique edge count	m̿ =	483,450,157
Wedge count	s =	42,676,090,519,343
Maximum degree	d_max =	2,323,263
Maximum left degree	d_1max =	436
Maximum right degree	d_2max =	2,323,263
Average degree	d =	176.925
Average left degree	d₁ =	89.984 0
Average right degree	d₂ =	5,231.52
Average edge multiplicity	m̃ =	1.526 26
Size of LCC	N =	8,341,043
50-Percentile effective diameter	δ_0.5 =	1.822 50
90-Percentile effective diameter	δ_0.9 =	3.731 23
Mean distance	δ_m =	2.764 15
Balanced inequality ratio	P =	0.283 092
Left balanced inequality ratio	P₁ =	0.413 586
Right balanced inequality ratio	P₂ =	0.102 089

[1]	Jérôme Kunegis. KONECT – The Koblenz Network Collection. In Proc. Int. Conf. on World Wide Web Companion, pages 1343–1350, 2013. [ http ]
[2]	M. Lichman. UCI Machine Learning Repository, 2013. [ http ]