Re: Resignation

> -----Original Message-----
> From: Next generation catalogs for libraries
> [mailto:NGC4LIB_at_listserv.nd.edu] On Behalf Of T Scott Plutchak
> Sent: Friday, August 31, 2007 4:23 PM
> To: NGC4LIB_at_listserv.nd.edu
> Subject: Re: [NGC4LIB] Resignation
>
> Both SCOPUS and Web of Science have implemented author disambiguation
> features in the past year, designed to do the kind of thing that Conal
> describes in order to determine which among authors of similar names (or
> variant names) is the right author for a particular article.  I don't
> have any sense of whether these approaches do the job better or worse or
> just as good as humans, but they are real-world applications currently
> being actively marketed.
>
> Here's a link to an Information Today article from July 2006 that
> describes them.
> http://newsbreaks.infotoday.com/nbreader.asp?ArticleID=16997
>
>
> T. Scott Plutchak

This is nice to see, and I hope it works. For the record, if we could get
rid of cataloging/subject indexing, and the final product were reliable,
that would be a huge gain for everybody. I personally think there is, and
will be, plenty of work for "information managers" within libraries and
without.

But I must stress again: these things need to be evaluated by disinterested
experts, not by the people inside. For example, in this article, it was the
director of marketing who said "99 out of 100 records are matched correctly
to an author with 95 percent recall; only five records would not be matched
due to insufficient data." This might sound pretty good until you get into a
discussion of what these numbers mean. I remember being similarly impressed
when I read abou the "accuracy" of optical character recognition (98%
accuracy or so), which sounds good until you see the final product.

Here is a sample taken at random from the Google Books project. This is from
Boswell's biography of Johnson.

"But checking himaelf and
1 As this subject frequently recurs in these volumes, the reader may be led
erroneously
to ¿¿ppose that Dr. Johnson was so fond of such discussions, as frequently
to introduso
them. But the tr.ith is, that the author himself delighted in talking
concerning gho*a,
and what he has frequently denominated the mysterious; and therefore took
every oppor.
tunity of leading Johnson to converse on such subjects—MALONE. "

Regards,
Jim

James Weinheimer  j.weinheimer_at_aur.edu
Director of Library and Information Services
The American University of Rome
via Pietro Roselli, 4
00153 Rome, Italy
voice- 011 39 06 58330919 ext. 327
fax-011 39 06 58330992