17 Nov
2012
17 Nov
'12
4:15 p.m.
On 17/11/12 6:21 PM, Fred Cisin wrote:
Yes, various aspects are published.
* http://infolab.stanford.edu/~backrub/google.html
* There is also a paper, "Page Rank Algorithm" from Dept of
Mathematics and Statistics, University of Massachusetts, Amherst, but
ironically Google's search results make it impossible for me to copy a
sane link, and when I click it, I don't get a copyable link in the
browser either.
--Toby
On Sat, 17 Nov 2012, Toby Thain wrote:
A lot of their endeavours build on the same infrastructure. BigTable,
BigFS, etc.
Yes, you are quite right, but you need the
endlessly scalable and
fault-tolerant architecture as well. So you could say two paradigm
shifts were involved. No wonder Google won. :)
Google has branched out into a lot of other Information fields.
I remember when they were frantically building a "war chest" in
expectation of MAJOR confrontations with Microsoft. Once you got used to the idea of the page you
want being #1 most of the
time (as is still the case IME), then one quickly forgot about the other
engines...
During their early days, Google did a surprisingly large amount of their
relevance ranking MANUALLY! Then they apparently expanded into letting
more of their ranking be alfgorithmically done with Saltonin vector Space
and cosine measures. I wonder what they are doing now. Have they
published any of their algorithms in any of the Information Science
journals?
--
Grumpy Ol' Fred cisin at xenosoft.com