The TrustRank Algorithm - Sounds like PageRank with Knobs on

5 comments
Thread Title:
TrustRank Algorithm
Thread Description:

There's a new white paper out that outlines the TrustRank [pdf] algorithm that some of you will find interesting no doubt.

It's written by Zoltan Gyongyi, Hector Garcia-Molina, and Jan Pederson

Here's a quote:

Human editors help search engines combat search engine spam, but reviewing all content is impractical. TrustRank places a core vote of trust on a seed set of reviewed sites to help search engines identify pages that would be considered useful from pages that would be considered spam. This trust is attenuated to other sites through links from the seed sites.

To me, that sounds like pagerank with knobs on, and just as simple to game - but then im not a harvard doc right? heh..

If the technically inclined would care to give their opinions on the paper it would be cool...

Comments

I'll go through the paper in

I'll go through the paper in more depth of the coming week or so but for now here are my first thoughts based on Very quick overview and gut feeling.

I thought I had read this before, but I wasn't sure so I did a little digging. It seems the paper was published on the 11th of March 2004 as can be seen from the Stanford Publication Server so this isn't new news.

Even though it is well worth a more thorough read as it throws up some interesting questions.

  • What is this paper's aim?
  • What does this paper mean to SEOers?
  • Is Trustrank being used today?

What is this paper's aim?

IMHO the paper's aim is simple. "It is to eliminate the ability to spam the Pagerank algo" It specifically says within that

Quote:
Our work builds upon existing PageRank research

and

Quote:
The term web spam refers to hyperlinked pages on the World Wide Web that are created with the intention of misleading search engines

What does this paper mean to SEOers?

It depends on how you currently work. If you are a white hatter, it might affect you if you aren't already included in some of the more major directories out there. What those directories are is open to debate but as Jan Pedersen, chief scientist at Yahoo is one of the paper's authors I believe it is fair to include the Yahoo directory.

If you wear a darker coloured hat then I doubt it will affect you in any shape or form. Carry on as normal, remembering that other algos look at relationships in a lot more detail, and if you can beat them then this will look more like a mini mound rather than the top of the hill

Is Trustrank being used today?

I don't believe it is being used in its current form, but I do believe that parts of it are incorporated into other algorithms that are in full force, and have been for some time!

Hm, not that new...

http://www.webmasterworld.com/forum34/630.htm

Back when I still read that site regularly ;) The idea seems sound to me, and Garcia-Molina's name is on a few other good search research papers...

Thinking that at some point - the research must be put to use, ya? Between this & other techniques of page analysis to determine where the links occur in the doc & weighting them accordingly...imho, the 'basic link algos' of today could get better.

Not new..

One of the drawbacks of rapid publishing im afraid, my bad - sorry heh..

Interesting regardless of age though...

Very interesting indeed

There are many applications of the trust rank paper, it mentions P2P systems and search - but I think you could extend it to RSS aggregation & a host of other things.

>>rapid publishing

Would rather see a repeat of important stuff here, than miss something critical because folks were afraid of pointing stuff out that had been covered elsewhere...

Google Blogscoped hits the subject

I gave them a few suggestions at their site and at mine [url=http://tomnorian.com/index.php/2005/04/25/trust_rank_and_can_google_engineers_urge ]link[/url] on what little old me sees the dangers and benefits associated for them to keep in mind.

Basically they've got incentives to not be too limmiting in seed selection and to avoid their nature of looking for a firm mathematical solution and black or white answers....I used a few more words than that but you'all might find some of it reinventing the wheel.

Not sure that any google people read the google blog but I can always try to get my opinion heard?

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.