citing/cited patterns Scientometrics in the SCI-JCR 2003

[Loet Leydesdorff]

Dear David, Howard, and colleagues,

Let me first answer David's question: the algorithm used in this
representation is from

Kamada, T., & Kawai, S. (1989), "An Algorithm for Drawing General Undirected
Graphs", Information Processing Letters, Vol. 31 No. 1, pp. 7-15.

It is contained in Ucinet, Netdraw, and Pajek which come as a package as
indicated by Howard in his previous email. Pajek is also stand alone and
then freely available (for academic purposes) at
http://vlado.fmf.uni-lj.si/pub/networks/pajek/ . It is made and maintained
by a number of mathematicians of the University of Ljubljana. Pajek has
become a standard in many applications also outside social network analysis
(e.g., in evolutionary economics), but it is fully interfaced with Ucinet
(like Netdraw). One can probably do approximately the same things with all
these programs.

Indeed, Howard, you have the first publication using this with cited/citing
in JASIST. My compliments! I used Pajek for the visualization in my analysis
of USPTO data in terms of scientific references contained in patents in
JASIST 55(11), 991-1001 and in another paper which is still under submission
(http://www.leydesdorff.net/stemcell ). In the latter paper, I use all
citations from one journal (Bone and Marrow Transplantation) in a year in
order to see how they operate (asymmetrically).

While playing with the update of the JCR 2003, it occurred to me that the
journal citation matrix can also be fed directly into this same routine for
bi-modal matrices. Robert Tijssen (Leiden) had done this using
Quasi-correspondence analysis in Scientometrics 11 (1987) 347-361 and I had
always been a bit jealous on his pictures because they capture both
structures (citing, cited) in a single picture while I had to use two
pictures (e.g., Scientometrics 9 (1986) 103-125). Another problem at that
time was the use of different similarity criteria and clustering algorithms
as discussed more recently in several contributions on this list.

Because one does no longer need a similarity criterion, the size of the
files is virtually unlimited. I suppose that this is major advantage of
Pajek (above Netdraw). Viv Cothen (at Wolverhampton) is using the program
for the visualization of hyperlinks in Internet research using very large
files. I tried to feed the whole JCR (2003) into one representation (5907
journals), but the structure can then no longer be insightful because of the
impossibility to have 5907 labels on a page. (Perhaps, I should do a paper
on this for Stockholm.) For smaller matrices, however, one obtains an
insightful view which may allow us to get more handles on citation as a
selective operation.

With kind regards,


Loet
  _____

Loet Leydesdorff
Amsterdam School of Communications Research (ASCoR)
Kloveniersburgwal 48, 1012 CX Amsterdam
Tel.: +31-20- 525 6598; fax: +31-20- 525 3681
 <mailto:loet at leydesdorff.net> loet at leydesdorff.net ;
<http://www.leydesdorff.net/> http://www.leydesdorff.net/


 <http://www.upublish.com/books/leydesdorff-sci.htm> The Challenge of
Scientometrics ;  <http://www.upublish.com/books/leydesdorff.htm> The
Self-Organization of the Knowledge-Based Society



  _____

From: ASIS&T Special Interest Group on Metrics
[mailto:SIGMETRICS at LISTSERV.UTK.EDU] On Behalf Of Howard White
Sent: Monday, September 13, 2004 3:01 AM
To: SIGMETRICS at LISTSERV.UTK.EDU
Subject: Re: [SIGMETRICS] citing/cited patterns Scientometrics in the
SCI-JCR 2003


David Post has asked Loet Leydesdorff what "program/algorithm"
he used to create his diagram of citing/cited patterns in Scientometrics.
I hope Loet will not mind if I point out an earlier appearance of a similar
diagram on p. 120 of:

White, Howard D. Barry Wellman, Nancy Nazer.  (2004).  Does
    Citation Reflect Social Structure?  Longitudinal Evidence from
    the "Globenet" Interdisciplinary Research Group.  Journal of the
    American Society for Information Science and Technology 55(2):
    111-126.

The software used to make the diagram is Netdraw, created by Steve
Borgatti and distributed as part of UCINET 6, the premier package for
analyzing social network data. It works beautifully with citation network
data as well. All one needs is a matrix of sender-receiver (e.g.,
citer-citee)
relations in proper UCINET format and a brief introduction to Netdraw.
My article above could be consulted for some descriptive details.

UCINet and Netdraw are available from Borgatti at:

    http://www.analytictech.com/


Howard D. White
College of Information Science and Technology
Drexel University
Philadelphia, PA 19104



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.asis.org/pipermail/sigmetrics/attachments/20040913/94d67463/attachment.html>