Pagerank is all about link worth. It has nothing to do with the content on the page, the relevancy or anything else unless they've changed it drastically. It's the aggregate score for a page derived from all the pages that link to it, and all the pages that link to those pages, etc. It's a popularity contest where the judge is a mathematical formula.
Is just my personal observation that pages with good content tends to gain PR easily. I may be wrong, but then no one can be 100% sure about Google algo for giving PR (except page and brin and few others in google's den)
DomainMagnate (17 December, 2008)
That seems a bit like believing in the Tooth Fairy, the Easter Bunny, and Santa Claus.
I've seen a lot of advice from webmasters and so-called seo experts suggesting that they 'know what Google wants', and have seen their own sites take a tumble.
Certainly link building is important. I don't doubt that is the major part of page rank. But I seriously doubt that either of you could say without a doubt that it is nothing but link popularity. If you can say it without a doubt, I'd suggest not waiting up for Santa Claus this Christmas Eve; you might be disappointed.
Then this must be the equivalent of taking a DNA sample from Mr. Claus and verifying his identity.
That is the original paper that Larry and Sergey did while students at Stanford. The PR calculation is described in it.
There's also a million other explanations of PR.
PR has nothing to do with determining keyword relevancy. It's simply a iterative vote. A way to calculate importance of a page relative to all the other pages indexed. Several different calculations are involved in determining which results are returned. PR in the end is one factor in determining ranking. But being seen as an important page on the web will help a lot. As its put in the paper, PR is used to "prioritize" results. In other words, PR isn't the matching algorithm, another algo or procedure is applied and then "prioritized" using PR.
Originally Posted by The Stanford Paper