7th Sep 2008
by Darrin J. Ward:
I've always been greatly interested in mathematics. Well, not always, but I did come to have a lot respect for applied mathematics and physics during my latter years of school and college. Now, I have to also admit that I don't understand as much as I'd like, because it would simply take far too much time to learn it all. The deep stuff is beyond me and I admit that. Nonetheless, I remain fascinated by the sheer logic in math and the fact that it transcends race, time, other languages, etc. It's a universal language
Ever since I learned about Fermat's Last Theorem, I've been absolutely engrossed by the notion that a simple-looking and simple-sounding statement could boggle the minds of the world's greatest mathematicians for over 350 years. The theorem states, simply, that xn+yn=zn has no solutions where x,y and z are integers greater than zero and n is an integer of value 3 or greater. You'll note that n=2 would be the pythagorean theorem!
So, where is all of this going and how does it relate to SEO? Well, in reading the amazingly complicated Proof of Fermat's Last Theorem [PDF] by Andrew Wiles (and yes, I've actually had a printed copy in my office for the last few years), I've been forced to learn a little bit about some intriguing things in number theory. One such thing was Eigenvectors. In doing further research on these I came across a wonderful paper entitled "The $25,000,000,000 Eigenvector - The Linear Algebra Behind Google" by Kurt Bryan & Tanya Leise, which is basically about Google's PageRank (an Eigenvector).
I've read quite a lot of academic papers that theorize on various thing, but I had not come across this particular one before, so it was a pleasure to look through it. I mostly use academic papers as a source of inspiration rather than a solid foundation for an SEO campaign. They are extremely wonderful in provoking me to think about abstract things which eventually help me get ahead in the SEO world.
The fact of the matter is that search engines are nothing more than big calculators (though, with an arguable component of manual reviewing, a-la Google's Patent # 7096214). If you know how they work and understand the steps that they make in performing their calculations, then you have a significant competitive advantage. Looking at what's being proposed in these academic papers therefore makes a lot of sense as they are a great source of the latest in terms of strategies.
So, here are some of the papers that I usually recommend to people wanting to learn more. They do have a lot of mathematics in some cases, but you can usually get some good info even without understanding everything (I will update this list every-so-often, Contact me with addition considerations):
Authoritative sources in a Hyperlinked Environment
-- by Jon. M. Kleinberg
Site Level Noise Removal for Search Engines
-- by Andre Luiz da Costa Carvalho, Paul-Alexandru Chirita, Edleno Silva de Moura, Pavel Calado, Wolfgang Nejdl (2006)
The Anatomy of a Large-Scale Hypertextual Web Search Engine
-- by Sergey Brin and Lawrence Page
A Survey of Eigenvector Methods For Web Information Retrieval
-- by Amy N. Langville & Carl D. Meyer
ParaSite: Mining Structural Information on the Web
-- by Ellen Spertus
The $25,000,000,000 Eigenvector - The Linear Algebra Behind Google"
-- by Kurt Bryan & Tanya Leise