How about feeding back from web request stats?
Things that get pulled more often are probably more popular.
It's admittedly not very clever, but it would be easy to implement...
Digital Research and Scholarship R & D
the University of Virginia Library
On Dec 7, 2009, at 5:56 PM, Ed Summers wrote:
> On Mon, Dec 7, 2009 at 3:43 PM, LeVan,Ralph <[log in to unmask]> wrote:
>> For VIAF, rankings are calculated based on the number of institutions
>> that have controlled that name and the amount of attention the
>> institutions have given to that name (e.g. size of their respective name
>> authority records).
> Neat. It would be great to have some external dataset to use in
> ranking LCSH suggestions at id.loc.gov. But at the moment it's a
> simple mysql db loaded up with some MARC LCSH data. I guess it could
> do something smart with PageRank-like ranking of 'super-concepts'
> (concepts that are linked to a lot)...but that would've taken longer
> than 20 minutes :-)