On Mon, Dec 7, 2009 at 5:56 PM, Ed Summers <e...@pobox.com> wrote: > 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 :-)
The frequency of an LCSH term within the LC catalog could also be useful for ranking, although I'm not sure if such data would be readily available. Another possibility would be a simple count of broader terms + narrower terms + related terms or something like that. Although PageRank would probably be better, since even some "important" terms might have a relatively small number of immediately-adjacent links. Keith