Academic papers on search engine rankings

Can't seem to find much.

[pgmason]

Hi All. This is my first post here so I hope its in the right forum...

I'm currently doing an MSc in Information management at the University of Sheffield, and my dissertation is going to be based around search engine optimisation.

I've been looking for any academic papers, since a dissertation is primarily meant to focus on academic literature, rather than real-world communications like forum posts.

There doesn't seem to be a lot out there about how rankings work and how to change them. There's quite a bit on user behaviour, and lots on evaluating and comparing different search tools, but it seems there's very little on how engines actually do their thing other than Brin & Page's original Google paper from '97 or so.

Am I just looking in the wrong place? Can anyone point me in the direction of some useful papers/conference proceedings/journal articles?

Cheers
Pete

[isitreal]

My suspicion is that the people who know what makes search engines really work are working for them. Since they are business organizations, what you are looking for is a published version of the fundamental intellectual property of a search company, you want to find published the guts of say google, which you actually already have, since you have the original google document.

This is sort of like looking for the code of the windows nt kernel in publically available formats. If you find something like this, be sure to post it here so I can read it.

Am I just looking in the wrong place?

No, you're not, forums like this is where the knowledge exists, it's where people who know this stuff hang out, just as with most web based technologies. The most advanced discussions of this stuff happens on the web, academic stuff will be written, is my guess, based on that work, but following it by a long enough stretch to make it academically acceptable. There is one open source search engine project called I think mozdex that may have publically available material, since it's open source.

If you read these threads very closely you will start noticing that some of the posters pretty much know how these search engines run, from long empirical experience, but since you can't cite a forum thread that won't help you much.

[msgraph]

[citeseer.ist.psu.edu...]
[citeseer.ist.psu.edu...]
[both have not been updated that much for a while but still, CiteSeer is one of the best places out there]

International WWW Conference Sites
[www2003.org...]
[www2002.org...]
[just keep going down from there, www(year).org]

[dbpubs.stanford.edu...]

[uspto.gov...]
[most of the good academic papers end up on the patent table]

[pgmason]

Thanks for that. Citeseer looks to have lots of good stuff - seems everyone and his brother's written a new search engine!

The patent office is a good idea.

Cheers
Pete

[Swash]

[cs.toronto.edu...]

great reading there.

[brotherhood of LAN]

try a few related searches for "Adversarial IR", some class lectures specifically mention this forum.

[ukgimp]

An old post by me, not far from my firstever I think.

Anyhow, there is a reply by me where I mention some articles and at the bottem a good book recommedation.

[webmasterworld.com...]

ps dont laugh at me trying to get my head round the indexing porcedure :)

[nuevojefe]

Try some searches with ".edu" in them. It's a good way to find less SEMish sites.