Reverse Engineering Google's Algos

Gophering has been stressing that he's not sure where this thread is going, it may well be that there is very little usable information we can find out, but it'll all be a good learning path. Nobody is saying that we'll be able to find the 'magic formula' to insta-rank any website #1, because... we won't. But it will provide an interesting insight into some of the functionality of Google, and even if a lot has changed you can bet some aspects will be similar to the current Google set up.
Click to expand...

I am aware JFoulds I am just saying. I know from my friends with programming skill how hard it could be to look into somebody else's code
 
Advertise on BHW
Amazing, this is beyond interesting, you sir, got a subscriber.
 
The most important items that we need to know are the penalty flags. What are the penalty triggers. We all know how to rank websites. We just need to know the penalty flags so that we do not trigger them and rank websites even better or faster.

We do not need to know the whole program or how the program works. We only need certain information.
 
dragonrage01 said:
The most important items that we need to know are the penalty flags. What are the penalty triggers. We all know how to rank websites. We just need to know the penalty flags so that we do not trigger them and rank websites even better or faster.

We do not need to know the whole program or how the program works. We only need certain information.
Click to expand...

Unless, of course, dodging all penalties is a penalty of itself...
 
Fucking amazing. Upload the files asap man. I for one will help you look over them.
 
Gophering said:
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?
Click to expand...

Learn and make an ebook. Secret, word of mouth marketing, hefty price and you're golden.
D.
 
use the appliance by creating html documents with different parameters and having it index them. and observe how it ranks.. But I think you are wasting your time. GSA will likely only get you a feel for on page factors.
 
Last edited:
dbyrn said:
Learn and make an ebook. Secret, word of mouth marketing, hefty price and you're golden.
D.
Click to expand...

I've heard WF charges for an account nowadays, and I'm not about to contribute to that bunch of shit.

use the appliance by creating html documents with different parameters and having it index them. and observe how it ranks.. But I think you are wasting your time. GSA will likely only get you a feel for on page factors.
Click to expand...

GSA bases quality on offpage as well as onpage factors as evident from the screenshots.
 
Quick update:
Unfortunately I'm a bit busy this week. I've finally extracted the files from the partition (this was quite a fucking quest in itself, as I didn't use the obvious approach, which I should have...)

I'm just gonna go through the files, remove unnecessary stuff and upload all of this somewhere. Probably in the form of a torrent. Should be up by tomorrow or so. Apologies for the delay and thanks for your patience.

Cheers
 
Gophering said:
Quick update:
Unfortunately I'm a bit busy this week. I've finally extracted the files from the partition (this was quite a fucking quest in itself, as I didn't use the obvious approach, which I should have...)

I'm just gonna go through the files, remove unnecessary stuff and upload all of this somewhere. Probably in the form of a torrent. Should be up by tomorrow or so. Apologies for the delay and thanks for your patience.

Cheers
Click to expand...

Sounds fantastic, the community appreciates all your effort!
Let me know once it's uploaded, and I'll seed ad infinitum.
 
For everyone who's interested in reverse engineering this code, I've created a public Github repo so we can all collaborate together. We can then post any interesting findings in here instead of cluttering up the thread with general discussion.

Please PM me and I'll add you to the collaborators list and give you the URL.
 
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",
    2: "PAGE_MAILING_LIST",
    3: "PAGE_BLOG_COMMENTS",
    4: "PAGE_BLOG",
    5: "PAGE_PPC",
    19: "PAGE_SPAM_SIGNATURE_KILL",
    21: "PAGE_SPAM_SIGNATURE_NOPROP",
    7: "PAGE_AFFILIATE",
    17: "PAGE_WIKI",
    1001: "PAGE_BLOG_COMMENTS_TMP_FOR_ALEXIS",
    1002: "PAGE_BLOG_COMMENTS_TMP_FOR_VINEET",
    8: "PAGE_MOBILE",
    9: "PAGE_NOFOLLOW",
    10: "PAGE_NOINDEX",
    11: "PAGE_HTTP_BAD",
    12: "PAGE_LINKSPAM",
    13: "PAGE_DEMOTE",
    14: "PAGE_HAS_FRAMESET",
    15: "PAGE_BADURLS_NOPROP",
    20: "PAGE_BADURLS_NOPROP_ALL",
    22: "PAGE_FRIENDLINK",

And links are split into the following categories:
Code: 
6: "LINK_PAID",
    18: "LINK_REFERRER",
    1: "LINK_FORUM_COMMERCIAL",
    64: "LINK_BOILERPLATE",
    65: "LINK_ADBANNER",
    66: "LINK_IMAGE",
    67: "LINK_EXPIRED_PURGATORY",
    68: "LINK_NOFOLLOW",
    69: "LINK_DEMOTE",
    70: "LINK_LINKDROP",
    71: "LINK_BEFORE_EXPIRATION",
    72: "LINK_BEFORE_CREATION",
    73: "LINK_AFTER_EXPIRATION",
    74: "LINK_IMPORTED",
    75: "LINK_AUTOLEAK",
    76: "LINK_NOSCRIPT",
    77: "LINK_SOURCE_BOILERPLATE",
    78: "LINK_ALL_LINKEXCHANGE",
    79: "LINK_SECTION_LINKEXCHANGE",
    80: "LINK_TRANSIENT_BOILERPLATE",
    81: "LINK_SPAM_SIGNATURE_ALL",
    82: "LINK_SPAM_SIGNATURE_SECTION",
    83: "LINK_FRIENDLINK",
    84: "LINK_FRIENDLINK_COMMERCIAL",
    85: "LINK_BACKLINK_CLUSTER",
    86: "LINK_OUTLINK_CLUSTER",
    87: "LINK_MACHINE_GENERATED",
    88: "LINK_EMBEDED",
    89: "LINK_USER_COMMENT_BOILERPLATE",
    90: "LINK_AFFILIATE",

Only just started looking through, but some really interesting stuff here!
 
Good stuff, good stuff. I'll join you in your efforts as soon as I get some time to (prolly next week).

Cheers
 
That's some very interesting stuff, though if we could find out how the "brain" determines how to classify a link, it would be even better.
 
Waiting for your torrent release :-)
Gophering said:
Quick update:
Unfortunately I'm a bit busy this week. I've finally extracted the files from the partition (this was quite a fucking quest in itself, as I didn't use the obvious approach, which I should have...)

I'm just gonna go through the files, remove unnecessary stuff and upload all of this somewhere. Probably in the form of a torrent. Should be up by tomorrow or so. Apologies for the delay and thanks for your patience.

Cheers
Click to expand...
 
Advertise on BHW
You must log in or register to reply here.
Sign up now!

Main Menu

Home Main Forum List Black Hat SEO White Hat SEO BHW Newbie Guide Blogging Black Hat Tools Social Media Downloads

Marketplace

Account Selling Content / Copywriting EBooks / Methods Hosting Misc Proxies For Sale SEO - Link building SEO - Packages Social Media Social Media - Panels Web Design More

Making Money

Affiliate Programs Hire a Freelancer IM Journeys Making Money Pay Per Click Site Flipping Want to Buy

BlackHat World

Dispute Resolution Forum Suggestions & Feedback Introductions Lounge
Back
Top