As a public service, I have decided to explain to people how to improve the pagerank of their site. First you need to understand what pagerank is. That is the easy part. PageRank corresponds to the principal eigenvector of the normalized link matrix of the web. It can be calculated using a simple iterative algorithm. The rest of the details can be found here. Once that is understood, then we can move on to how to increase pagerank.
14 thoughts on “How to improve your pagerank.”
Comments are closed.
Thanks for that simple explanation!
And yeah, the link.
Ravi, you should author this segment with Arindam Chaudhuri. He seems to have found a magic way to increase pagerank.
Why go through all the trouble of learning about eigenvectors and what not, when you can look up the papers for an institute that provides you with a certificate that says you are a pagerank holder?
While doubting Page, Brin and Google would only expose my ignorance, I still think this area has not seen as many results as one would have liked.
This paper was a path breaking one. But I expected lot more contributions based on this technique. For example, the principal eigen vector has some severe restrictions whn taken in isolation. It does not give any special weightage to say, a cult like following.
The results on correlations between traffic and page rank have also left a lot to be desired. I think this would be an ideal project for a graduate student – it is a solvable problem using standard engineering math, while it still is no walk in the park.
Well Nilu, if you insist on turning this into a serious discussion 😉
I’d expect that the questions you are raising would be the subject of continuous research by Google’s army of PhDs. I don’t think that the problem you are raising yields to a mathematically elegant solution. It is a question of heuristics. If google has developed techniques to penalise link farms, I suppose it would have developed techniques to penalise cultlike followings too..
Correlation between traffic and pagerank, on the other hand might have a mathematical solution. The problem would be getting accurate data. I’ve heard that the google toolbar is a way to obtain data about visits and stuff.
I think that the most difficult problem for google would be factoring out the google effect. The fact that google itself can affect traffic, linkages and pageranks ought to be a major problem for it. After all, the paper assumes that the search engine is just a disinterested observer in the whole system.
I don’t even remember what an eigen vector is 🙁
Lakshmi, I will pretend not to know what you majored in 😉
Man…this logic works believe me….Now if you search for “Funny Blog” in Google..I get my Blog first! Yahoo!
I understand the first part does not have a clear solution. But I would have expected some serious additions in the past 10 years.
I am sure there have been lots of additions in the past 10 years, all of them unfortunately proprietary and hidden within the Googlplex 🙁
If you are legit then call it improving page rank, otherwise its called “google bomb”.
RaviKiran : Is there a way to contact you? I wanted to invite you on our Podcast. Your reference from Patrix.
That’s a nice, comprehensive post that you have got there, Ravikiran. I hope that you also have a post for newbie bloggers on practical tips and do’s and don’ts. For a newbie everything is an asset and things like RSS etc are greek. Nice of you to share information. Thanks.
DOH
My primary interest is in cloaking… the good and bad. I have used it, but never found any benefit or penalty.
Would love you hear any and every pro and con, or advice in general.
——————————-
Search Engine Marketing Software… Super Affiliate