webkb — WebKB graphs (1997)

Description

Networks of hyperlinks among web pages of the computer science departments of four U.S. universities (Cornell, Texas, Washington and Wisconsin), from the WebKB project of Carnegie Mellon University, which collected the pages in January 1997 and manually classified them by their content. This version was prepared for the NetKit-SRL toolkit of S. A. Macskassy and F. Provost, and includes only the pages with one of six classes: course, department, faculty, project, staff, or student (the pages of the class "other" are excluded). Nodes are web pages. For each university there are two networks:

• webkb_<university>_link1: directed network where an edge $i \to j$ indicates that page $i$ contains hyperlinks to page $j$.
• webkb_<university>_cocite: undirected co-citation network, where two pages are connected if they are both linked from at least one common page.

The networks have the following properties:

• name: the identifier of the page (e.g. url-course-cornell-00002), which includes its class.
• weight: in the link1 networks, the number of hyperlinks from page $i$ to page $j$; in the cocite networks, the number of pages that link to both $i$ and $j$.
• class: the class of the page (course, department, faculty, project, staff or student).

Tags

Informational Web graph Weighted Metadata

Citation

• S. Slattery & M. Craven, "Combining Statistical and Relational Methods for Learning in Hypertext Domains." In Proc. 8th International Conf. on Inductive Logic Programming, 38-52 (1998) 10.1007/bfb0027309 Sci-Hub

Upstream URL OK

http://netkit-srl.sourceforge.net/data.html

Networks

Tip: click on a column header to sort the list, or hover over it for a description.

Column legend

$\left<k\right>$

Mean degree

$\sigma_k$

Standard deviation of the degrees

$\lambda_h$

Largest eigenvalue of the Hashimoto (non-backtracking) matrix

$\tau$

Random walk mixing time

$r$

Degree assortativity

$c$

Global clustering coefficient

$\oslash$

(Pseudo-) diameter

$S$

Relative size of the largest component

NPs

Node properties

EPs

Edge properties

Show all columns

Name	Nodes	Edges	$\left<k\right>$	$\sigma_k$	$\lambda_h$	$\tau$	$r$	$c$	$\oslash$	$S$	Kind	Mode	NPs	EPs	gt	GraphML	GML	csv
webkb_wisconsin_link1	300	1,155	3.85	10.27	11.97	26.35	-0.10	0.19	11	0.93	Directed	Unipartite	name student	weight	8 KiB	12 KiB	12 KiB	9 KiB
webkb_wisconsin_cocite	348	33,250	191.09	137.91	154.99	3.81	0.41	0.83	4	1.00	Undirected	Unipartite	name student	weight	22 KiB	108 KiB	107 KiB	89 KiB
webkb_washington_link1	433	1,941	4.48	18.18	12.97	29.97	-0.22	0.04	9	1.00	Directed	Unipartite	name student	weight	11 KiB	17 KiB	17 KiB	14 KiB
webkb_washington_cocite	434	30,462	140.38	131.75	139.94	4.75	0.25	0.76	4	1.00	Undirected	Unipartite	name student	weight	27 KiB	109 KiB	111 KiB	85 KiB
webkb_cornell_link1	349	1,393	3.99	16.88	8.33	11.28	-0.22	0.03	8	0.98	Directed	Unipartite	name project	weight	9 KiB	14 KiB	13 KiB	11 KiB
webkb_cornell_cocite	346	26,832	155.10	122.23	133.35	2.97	0.35	0.79	4	1.00	Undirected	Unipartite	name project	weight	20 KiB	92 KiB	93 KiB	70 KiB
webkb_texas_link1	286	1,002	3.50	12.87	7.45	30.68	-0.22	0.07	10	0.94	Directed	Unipartite	name faculty	weight	7 KiB	11 KiB	11 KiB	9 KiB
webkb_texas_cocite	334	32,988	197.53	135.58	152.98	8.59	0.41	0.81	5	1.00	Undirected	Unipartite	name faculty	weight	19 KiB	109 KiB	107 KiB	87 KiB

Ridiculograms^*

^* These are automatically generated force-directed visualizations, and can be quite meaningless for networks both big and small. They should not be taken seriously as sources of scientific insight. See here for a discussion.