Forum Moderators: open
1) Google sorts pages based on content relevance as it always has.
2) Google then goes through that set of results and checks for "interconnectivity" between them.
3) Documents that are linked to by other documents in the original set of results are given a "local rank" value.
4) Multiply original rank times local rank (and a few other constants and threshold factors) and you've got a "Modified Rank".
5) Send the original results, but in the "Modified Rank" order.
Any info whould be great :)
_________
The method of claim 6, wherein refining the relevance score values for the documents further includes:
recalculating the relevance score values for the documents as
NewScore(x)=(a+LocalScore(x)/MaxLS)(b+OldScore(x)/MaxOS)
where NewScore is the recalculated relevance score value of the particular document, a and b are predetermined constants, MaxLS is equal to the maximum of the calculated local score values, MaxOS is equal to the maximum of the calculated relevance score values, and LocalScore(x) refers to the local score value of the particular document.
_____________
you can change the MaxLS and MaxOS and A,B to get new results.
I do believe that they are using this algo, it makes so much sense.
They only do it on the first 1000 results and the rest are the same.
Easy to do on the fly other engine do.
This alog checks for same content, inter-links (local), same host.
I believe it now important to get links from sites that are about the same content as you, which is VERY hard.
Plus the Patent talks about how easy it to play with the result.