NIPS full papers

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

Metadata

Code		`NI`
Internal name		`bag-nips`
Name		NIPS full papers
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 =	13,875
Left size	n₁ =	1,500
Right size	n₂ =	12,375
Volume	m =	1,932,365
Unique edge count	m̿ =	746,316
Wedge count	s =	316,414,350
Claw count	z =	60,669,295,416
Cross count	x =	10,271,177,488,043
Square count	q =	7,325,274,840
4-Tour count	T₄ =	59,870,386,232
Maximum degree	d_max =	1,455
Maximum left degree	d_1max =	914
Maximum right degree	d_2max =	1,455
Average degree	d =	278.539
Average left degree	d₁ =	1,288.24
Average right degree	d₂ =	156.151
Fill	p =	0.040 205 6
Average edge multiplicity	m̃ =	2.589 20
Size of LCC	N =	13,875
Diameter	δ =	6
50-Percentile effective diameter	δ_0.5 =	3.028 37
90-Percentile effective diameter	δ_0.9 =	3.805 67
Median distance	δ_M =	4
Mean distance	δ_m =	3.219 19
Gini coefficient	G =	0.821 596
Balanced inequality ratio	P =	0.140 265
Left balanced inequality ratio	P₁ =	0.454 947
Right balanced inequality ratio	P₂ =	0.178 579
Relative edge distribution entropy	H_er =	0.892 567
Power law exponent	γ =	1.300 92
Tail power law exponent	γ_t =	1.521 00
Degree assortativity	ρ =	−0.042 518 0
Degree assortativity p-value	p_ρ =	1.267 82 × 10⁻²⁹⁵
Spectral norm	α =	1,943.24
Algebraic connectivity	a =	0.996 481
Spectral separation	\|λ₁[A] / λ₂[A]\| =	2.387 05
Controllability	C =	10,875
Relative controllability	C_r =	0.783 784

[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 ]