It's not really a "new" form of citation analysis. There have been analyses
of co-circulation networks of books in libraries, which, along with
co-citation networks, can be used to help prioritize collection development
or retention. Also, bootstrapping a book recommendation system (as an
interface to a library catalog) can be done by loading in some recent set of
co-circulation data, and then using neural network processes to strengthen
links that are accreted, and weaken links that are ignored. This was the
basis of the Adaptive Library Network, which Paul Kantor and I got Dept of
Education funding to study in the early 90s. While we (and other)
researchers were struggling along with distributed PCs with the software to
interface with the University library catalog, Amazon and others were
implementing commercial recommender systems, much faster and better.
Ronald E. Rice
Professor, Chair of Department of Communication
School of Communication, Information & Library Studies
4 Huntington St., New Brunswick, NJ 08901-1071
w: 732-932-7381; f: 732-932-6916; e: [log in to unmask]
home page: http://scils.rutgers.edu/~rrice
----- Original Message -----
From: "Barry Wellman" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Wednesday, August 01, 2001 8:30 PM
Subject: new analysis
> I just realized that who buys whose books with who else's books is a new
> form of citation analysis.
> And it's asymmetrical, as I found out sadly. People who bought my books
> also bought Wasserman, NanLin, Lesser, etc (list of top 5 co-purchasers).
> But people who bought their books aren't listed among the top 5 of buying
> my books.
> Barry Wellman Professor of Sociology NetLab Director
> [log in to unmask] http://www.chass.utoronto.ca/~wellman
> Centre for Urban & Community Studies University of Toronto
> 455 Spadina Avenue Toronto Canada M5S 2G8 fax:+1-416-978-7162