Hi Rajesh and Friends,

Yes, Google Pigeon Rank™ technology description had lot of pun. It is only a different way of ranking web pages and nothing to do with Pigeons cooing!

The Pigeon Rank [Page Rank] method uses the link structure of web pages in World Wide Web to calculate a quality ranking for every web page. Google has created graph maps of a majority of the hyperlinks on the web, and based on the search hits, provides an objective measure of the citation importance, thus prioritize the web page titles [Parallels with Analytic Hierarchy Processing, AHP method developed by Thomas L. Satty]. Pigeon Rank [Page Rank] uses some mathematical calculations, a description of which is available at this link:
http://www-db.stanford.edu/~backrub/google.html
The more technically inclined can view the original paper on Page Rank at this link:
http://citeseer.nj.nec.com/page98pagerank.html

The calculations, historical index data, and ranking are done at a fast pace using the computing power [coops!] available with Google. The Pigeon Metaphor is used to exemplify the superior mental rotation, and independent generalization of recognition behavior of Pigeons when trained to discriminate between objects. Pigeons seem to rotate a mental image faster than the humans. Humans, unlike Pigeons use a view combination process, as we see patterns in total. This is a classic example where research in Psychology has helped a totally different domain.

Rajesh, regarding your question on how the method is different from the machine-learning technique, the authors have argued and proved in their research paper that;

1. Page Rank blends qualitative and subjective ranking that is essential for giving meaningful search results.

2. Page Rank uses an iterative algorithm for recognizing the patterns and is fast using lesser computing resources. A PageRank for 26 million web pages can be computed in a few hours on a medium size workstation

3. Improved scalability, low cost of indexing web page titles, and adaptability to the growth of www.

4. It models user behavior and has weights for unbreakable links, font sizes, and citations.

Some interesting papers written by Google employees can be accessed at http://labs.google.com/papers.html

Well there are critics for Google. There is a google haters club at http://www.google-watch.org/ , which argues that Google’s Page Ranking is biased and manipulated. Well, the world is always full of pessimists!

Sergey Brin and Lawrence Page through this small idea developed as Stanford students show what it takes to launch a big IPO.

Good Luck
Madan