Forum Moderators: open
I asked the same question on an other forum, and I got the following links :
[touchgraph.com...]
Looks like a better tool to analyse the similar pages. Reuiqres java plugin, and a bit of time (takes ages to download!)
Also, look at this previous discussion. [webmasterworld.com]
This is still a bit blur ... - the pages listed haven't got :
1) the same html structure (doesn't seem to be relevant any way if you read the other discussion).
2) None of the "backward links" actually link to the pages listed
3) None of the listed link to our page.
I am a bit confused ....
> 1) the same html structure (doesn't seem to be relevant any way if you read the other discussion).
Yes, I agree. There is no relation to the html structure.
> 2) None of the "backward links" actually link to the pages listed
> 3) None of the listed link to our page.
I didn't say that there is a direct connect between the pages. I think that if you are looking for "Similar pages" for page A you will get a list of pages B1, B2, ... because there are pages C1, C2, ... which link to both of them.
For example, related:www.google.com will show you Yahoo, Altavista, ... because there are a number of pages on the web, where you find links to search engines.
Of course, there is a complex algorithm to filter out the most important related sites.
If wou want to find a relation between www.example.fr and the pages shown for related:http://www.example.fr I would search for pages which contain both domains and look if there are links to both sites. I wouldn't use the "link:" command, because Google just show the important pages.