Re: How to compare research impact of toll- vs. open-access research

From: Matthew Cockerill <matt_at_biomedcentral.com>
Date: Tue, 27 Apr 2004 10:02:47 +0100

On Mon, 26 Apr 2004, Jim Till wrote:

> http://www.hql.com/content/1/1/16

Note: the URL link given above contains a typo - it should be
http://www.hqlo.com/content/1/1/16

> I have no credible explanation for the almost-immediate high rank that
> the Google page ranking algorithm gave to the CogPrints version.

Actually, if you use the Google toolbar to see the Pagerank of the article
directly, you can see that the Google pagerank is actually low (0/10) for
the article in Cogprints (since there hasn't been time for many links to it
to be picked up by Google)

But the pagerank of the specific URL is only one of the factors that Google
uses when ranking search results:

See section 4.5.1 The Ranking System of
http://www-db.stanford.edu/~backrub/google.html
for some details of how Google ranks results (though bear in mind that
Google continue to evolve the ranking function to work around abuses etc.)

Basically, factors such as the occurence of the keywords in the title make a
difference, as do the proximity of the occurences of those words, and even
the font size of the words in the document.

Also, my understanding is that Google makes use of other factors such as how
short is the URL of the page (the shorter the better), and how highly ranked
is the website on which it appears (this ensures that even a brand new page
on an important site will tend to be listed highly in Google results, even
though, technically, its pagerank is still zero).

Matthew Cockerill Ph.D.
Technical Director, BioMed Central
Received on Tue Apr 27 2004 - 10:02:47 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:27 GMT