Reverse Engineering Google's Algos

Waiting for your torrent release :-)

Tarball with library/source files was released above. More to come soonish. Contact JFoulds if you are interested in participating, as he's pushed it all to a public repo for easy access.

I'll get more stuff up as soon as possible.
 
I would like to play with this, can you post the download link?
 
We can see here (segindexer/anchors_pb.py) that Google classifies pages and links and then uses these to determine the quality of links/pages etc.

We can see here that Google uses algorithms to determine pages in the following categories:

Code:
 _LinkTags_NAMES = {
    0: "PAGE_GUESTBOOK",
    16: "PAGE_FORUM",
....

Only just started looking through, but some really interesting stuff here!

This is just awesome. If we would get the ranking algorithm for a page running as a standalone prog/script, this could be included in the usual SENuke's, GSA SER's etc. as last step to decide if a post/article is worthwile.

In any case if you get it running I suspect you could bank hard with it, i.e. as an addon to the aforementioned progs.
 
Its brains like this that make BHW interesting, period. Good luck I dont think you are wasting a single second.
 
Wouldn't have the slightest idea how to monetize this tbh, apart from somebody hiring me to do some coding based on my skillset displayed here (although next to no coding was done). You got any suggestions?

Not sure unless you get an inside track to the secret sauce and farm yourself as a Google expert. Maybe even author a WSO titled,"Cracking the Google Secret Sauce!" or actually a self-titled blog of the same name with your findings.
 
Could you run any a website through your little Frankenstein and see how it classified your existing pages and or links? I'd pay for that..
 
Not sure unless you get an inside track to the secret sauce and farm yourself as a Google expert. Maybe even author a WSO titled,"Cracking the Google Secret Sauce!" or actually a self-titled blog of the same name with your findings.

Appreciate the advice, however I doubt this could work as, in regards to Google, I haven't really cracked anything and I highly doubt I ever will.

Could you run any a website through your little Frankenstein and see how it classified your existing pages and or links? I'd pay for that..

Not a bad idea. A google sandbox environment. I might set up something like that just for the fun of it. Charging for this would be rather unethical though, as I can't guarantee Google's true behaviour.
 
Thanks mate. Better to keep a repo. Participating!! :)
Tarball with library/source files was released above. More to come soonish. Contact JFoulds if you are interested in participating, as he's pushed it all to a public repo for easy access.

I'll get more stuff up as soon as possible.
 
@OP: Thanks for this thread, great job man. Can't wait to dig my hands into the code! Even if Google's not using this code in their primary search algorithms it still I think will help to give us all some insight on how they think, which can be gold in of itself.
 
PM sent to JFoulds :-)
EDIT: I am also interested in the appliance. Can somebody kindly PM me the download link? :-)
 
wow, very nice and interesting! Can somebody PM me the download link? thx a lot :)
 
Thank you OP for starting this...I hope we all learn something that is helpful! PM'd JFoulds and hoping for a donwload link to the original file as well.
 
Great job man, really very excited to know more and thanks a lot for your hard work. Hope for the best :-) +rep added..
 
Great job but nothing useful so far.
We need to determine the penalties. How the penalties are being handled.
 
Back
Top