On 10 Mar , at 19:26:39, Tom Steinberg wrote:

> Name matching White House visitors and registered lobbyists:
> 
> http://sunlightlabs.com/blog/2010/lobbyists-and-white-house-visitors/
> 
> There must be lots of false positives, I guess, but an interesting idea.
> 

I just posted this:

Possible refinement, to address the "John Adams in one dataset may be a 
different John Adams in another" issue. If you can combine these data with a 
dataset that reflects the frequency of occurrence of that name in the wider 
population, you could come up with a confidence score for the uniqueness of the 
name. As a rough approximation "number of hits on facebook" might suffice;

compare:

http://www.facebook.com/search/?q=john+adams&init=quick

http://www.facebook.com/search/?q=stefan+magdalinski&init=quick

If I visited the White House, you can be fairly damn sure it isn't another 
Stefan Magdalinski


> Tom
> 
> _______________________________________________
> Mailing list [email protected]
> Archive, settings, or unsubscribe:
> https://secure.mysociety.org/admin/lists/mailman/listinfo/developers-public

--
/*
Stefan Magdalinski
+27 82 0431230 (phone)
smagdali (IM/twitter/flickr/dopplr/skype/etc)
*/


_______________________________________________
Mailing list [email protected]
Archive, settings, or unsubscribe:
https://secure.mysociety.org/admin/lists/mailman/listinfo/developers-public

Reply via email to