Jump to content

  • Log in with Facebook Log in with Twitter Log In with Google      Sign In   
  • Create Account

Subscribe to HRA Now!

 



Are you a Google Analytics enthusiast?

Share and download Custom Google Analytics Reports, dashboards and advanced segments--for FREE! 

 



 

 www.CustomReportSharing.com 

From the folks who brought you High Rankings!



Photo
- - - - -

Semantic Search


  • Please log in to reply
20 replies to this topic

#1 wendell

wendell

    HR 2

  • Members
  • PipPip
  • 19 posts

Posted 14 February 2010 - 12:18 PM

I searched around the forums for any relavent information on the latest semantic search offerings. Does anyone have an opinion of how a semantic search engine compares to a larger search engine like Google and Bing? Is anyone out there making an extra effort to rank well specifically in semantic search engines?

#2 Jill

Jill

    Recovering SEO

  • Admin
  • 33,244 posts

Posted 14 February 2010 - 12:33 PM

What makes you think Google, Bing and Yahoo aren't semantic search engines?

#3 PatrickGer

PatrickGer

    HR 5

  • Active Members
  • PipPipPipPipPip
  • 285 posts

Posted 14 February 2010 - 01:48 PM

I was gonna say: Nobody really should want to rank in search engines other than Google & Bing (in the US, UK, etc. SERPs that is), because a ranking in those would only bring a fraction of the traffic & business that a good ranking in Google & Bing can bring...as they have most of the search traffic conducted (in the US, UK, etc.).

But, I didn't say it (or well now I did lol), because it reminded me of something else:

The vast majority of searchers seem to click on the #1 ranking in the SERPs. I remember a blog post by Dazzlin' Dona (sp?) about that AOL data leakage...that basically 30% of the searchers click the #1 ranking. And then the CTR declines fairly quickly.

I'm not sure what that AOL data leakage was really all about, and averages are often useless (e.g. in niches where high-ticket items are sold people might do more research, not just click on a #1 ranking and buy), but something tells me there's some truth to that - as even I still find myself doing that way too much - clicking on the #1 ranking automatically.

Considering literally everyone's primary goal is to rank in Google ("If I optimize for Google I'll get a decent ranking in Yahoo, MSN, ... (now Bing, of course) anyway", maybe shooting for a #1 ranking in other SEs by optimizing specifically for another SE that brings decent traffic might be a highly underused, and thus profitable technique?

What I'm really trying to say is:

Another search engine might only bring in say 30% of the traffic Google can bring (no idea of Bing's market share to be honest), but maybe the much higher CTR of a #1 ranking (as opposed to a #2 or #3 ranking) might more than offset that, in some cases? and might be fairly easy to accomplish (in some cases, where you already are on page #1) considering virtually nobody optimizes for SEs other than Google these days?

Say you're working on a large site maybe you could go through all of your non-#1-but-top5-rankings (in G), have a look at the SERPs for them and find some nuggets, where a bit of bing-specific on-page optimization might be all it takes to go from #2-#5 to #1 - and get a much higher CTR (or only go through all of your #2 rankings as you might waste less time and get a higher ROI out of it).

Then again, maybe simply tweaking your titles for CTR's is usually a better use of one's time in most cases?

PS: Unfortunately, I dont have any idea if the #1 ranking maybe has a much lower conversion rate than a #5 ranking in most cases (people who click on the #1 ranking mindlessly w/o doing further research might be less likely to buy than someone who made the effort to check out multiple pages in the SERPs - including a #5 ranking for example, which then might have a higher conversion rate)

Nor have I done extensive research to understand how CTRs between search engines differ (though I know a guy who said he had a horrible c onversion rate for Google on one of his sites/products, and got most of his business from yahoo, b/c of a much better conversion rate)

#4 wendell

wendell

    HR 2

  • Members
  • PipPip
  • 19 posts

Posted 14 February 2010 - 02:18 PM

QUOTE(Jill @ Feb 14 2010, 12:33 PM) View Post
What makes you think Google, Bing and Yahoo aren't semantic search engines?


Jill: Yep, I worded my question wrong. Of course Google, MSN and Bing all use semantic techniques to present their results.

What I meant to say is, are there individuals that are trying to rank well within Hakia, Sensebot, Powerset and the other purely semantic search engines?

The reason why I'm asking is because I had a bit of a debate with another SEO professional about these (Hakia, Sensebot and others) sort of search engines. What I indicated to him was that the major search engines already use semantics to present their results so it's not necessary (as of today) to tailor a marketing plan that is going to get you a high ranking on Hakia if very few people actually use Hakia.

Patrick: Thanks for the information, very helpful.



#5 Jill

Jill

    Recovering SEO

  • Admin
  • 33,244 posts

Posted 14 February 2010 - 03:58 PM

QUOTE
What I meant to say is, are there individuals that are trying to rank well within Hakia, Sensebot, Powerset and the other purely semantic search engines?


Since nobody seems to use those engines, I don't know why it would be worthwhile to optimize for them.

That said, I would imagine that optimizing for Google would get you in good with those engines as well since I'm sure they want the same thing--the best pages for the search query at hand.

#6 Michael Martinez

Michael Martinez

    HR 10

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 5,325 posts
  • Location:Georgia

Posted 16 February 2010 - 06:14 PM

I have yet to find a semantic search engine that can really address the subject of dogs, canines, and man's best friend in the way I would expect a semantic search engine to do so.

As far as I'm concerned, it doesn't matter how semantic any search engine is. They have a lot of work left to do.


#7 PatrickGer

PatrickGer

    HR 5

  • Active Members
  • PipPipPipPipPip
  • 285 posts

Posted 16 February 2010 - 08:44 PM

@Michael Martinez: Very interesting to read what you say about this (considering you're apparently big on sticking to the facts), as this is a topic I've caught myself wondering about at the back of my mind before more than once (but dont really have the time to learn about this and research it at the moment having to balance everything with being in college).

Do you have any idea if the situation of machine translation is similar? Before I had started college, I considered majoring in "computational linguistics" - and read about the progress of machine translation that happened a lot slower than everyone (including people in the field) had expected it to, and that even the experts in the field dont think machine translation can substitute human translators anytime soon. Of course, this is only something I read about, thus the author might have guessed or lied himself, but considering that people seem to have been working on this stuff for a while and the best thing we have right now is babelfish (or do companies have better paid-for stuff?), and machine translation tools seem to be made mostly in order to help translators instead of substituting them...it made me think that its highly unlikely that machines might be able to do perfect human translation anytime soon (on the web).

Then fast forward a few years I hear in the Internet Marketing/SEO-field that computational power (or w/e it was they said exactly), etc. increased so quickly that those guys were expecting great machine translation to happen in the somewhat foreseeable future (decades)....dont really know what it was exactly anymore(sorry), but it was something along those lines. Actually, one person who I consider pretty smart said something along those lines, too.

Then I read a comment by fantomaster(hope i got his name right) saying something similar (about machine-translation) as what you just said about the state of semantic search.

Is there any reason to believe that the increasing growth of computational power (I feel like I might be confusing something, as I dont know exactly what they said - but it was about the expected explosion of computer stuff :-)) will help computers/machines become a lot better at translation?

Now, that I'm typing all of this..I cant help but wonder if that's a bit of an analogy to the ever-growing amount of data people have which doesnt necessarily help them make better decisions..or can even be counterproductive (if they already had enough data to begin with).

QUOTE(Michael Martinez @ Feb 17 2010, 12:14 AM) View Post
I have yet to find a semantic search engine that can really address the subject of dogs, canines, and man's best friend in the way I would expect a semantic search engine to do so.

As far as I'm concerned, it doesn't matter how semantic any search engine is. They have a lot of work left to do.



#8 chrishirst

chrishirst

    A not so moderate moderator.

  • Moderator
  • 7,718 posts
  • Location:Blackpool UK

Posted 18 February 2010 - 03:55 AM

mechanical translation -> http://www.webmaster...html#post949536

"Semantic search" is another nice idea in theory, but until we get to the point where English and American are the same language it is hardly likely to happen.



#9 1dmf

1dmf

    Keep Asking, Keep Questioning, Keep Learning

  • Active Members
  • PipPipPipPipPipPipPip
  • 2,167 posts
  • Location:Worthing - England

Posted 18 February 2010 - 05:50 AM

The likelyhood of every country in the world all speaking the same language and meaning the same thing is about as likely as me winning the nobel peace prize and the end of the world happening on the 21st December 2012, which is the end of the Mian calendar!

Or you could wait a 100 years.... English will turn into Panglish in 100 Years

Edited by 1dmf, 18 February 2010 - 06:48 AM.


#10 chrishirst

chrishirst

    A not so moderate moderator.

  • Moderator
  • 7,718 posts
  • Location:Blackpool UK

Posted 18 February 2010 - 09:59 AM

I don't know about 100 years, it's already happened in some towns. biggrin.gif





#11 PatrickGer

PatrickGer

    HR 5

  • Active Members
  • PipPipPipPipPip
  • 285 posts

Posted 18 February 2010 - 10:51 AM

QUOTE
English as it is spoken today will have disappeared in 100 years and could be replaced by a global language called Panglish, researchers claim.


"could", "researchers claim",...

Wait - it is not even "scientifically proven" (with a biased sample)?? Then I wont buy into it wink1.gif

EDIT:

Wow..just deleted my original post with the stuff above lol.

I also mentioned that when I was thinking of learning Chinese (was big into foreign languages before entering college and wanted to do something related to languages), people would say it doesnt make sense b/c it 10 yrs everyone in China will be fluent in English, they're all catching up, etc. ... One only needs to look at Japan, one of the richest and hardest-working nations in the world (where the state of English is said to be pretty much horrible (based on friends I know who lived and travelled there) to see it might not be that easy (they probbaly only have 24 hours in a day, too)

Edited by PatrickGer, 18 February 2010 - 11:05 AM.


#12 netometry

netometry

    HR 1

  • Members
  • Pip
  • 1 posts

Posted 18 February 2010 - 07:44 PM

"I was gonna say: Nobody really should want to rank in search engines other than Google & Bing (in the US, UK, etc. SERPs that is), because a ranking in those would only bring a fraction of the traffic & business that a good ranking in Google & Bing can bring...as they have most of the search traffic conducted (in the US, UK, etc.)." PatrickGR

"Since nobody seems to use those engines, I don't know why it would be worthwhile to optimize for them." Jill

Well, one reason to make sure you rank well within smaller search engines is the fact that one or two of them may become very important in the near future. Topsy, for example, though not a 'semantic' delivery, will probably get big if twitter keeps pushing them. Another is that if you rank well within the major and minor engines your personal interpretation of retrival algorithms is more generally positive. That's what an optimist should want, a general approach that works well in every circumstance. I've found that focusing on ranking well within smaller information services always brings about gains in google but not visa versa. Optimizing for only google is a bad idea but it probably has it's own topic... if i find it i'll post.





#13 1dmf

1dmf

    Keep Asking, Keep Questioning, Keep Learning

  • Active Members
  • PipPipPipPipPipPipPip
  • 2,167 posts
  • Location:Worthing - England

Posted 19 February 2010 - 05:22 AM

QUOTE
"could", "researchers claim",...

Wait - it is not even "scientifically proven" (with a biased sample)?? Then I wont buy into it
To be honest, who cares anyway, in 100 years i'll be dead so dont give a what language everyone will be speaking wink1.gif

Maybe we'll all evolve into vulcans and communitcate via telepathic mind melds! alien.gif

#14 chrishirst

chrishirst

    A not so moderate moderator.

  • Moderator
  • 7,718 posts
  • Location:Blackpool UK

Posted 19 February 2010 - 06:51 PM

QUOTE(1dmf @ Feb 19 2010, 10:22 AM) View Post
Maybe we'll all evolve into vulcans and communitcate via telepathic mind melds! alien.gif

Ah, but the mind meld does require physical contact,

Evolving into Betazoids would a better option biggrin.gif


#15 1dmf

1dmf

    Keep Asking, Keep Questioning, Keep Learning

  • Active Members
  • PipPipPipPipPipPipPip
  • 2,167 posts
  • Location:Worthing - England

Posted 21 February 2010 - 07:44 AM

QUOTE
Ah, but the mind meld does require physical contact,

Evolving into Betazoids would a better option


Not really, at least requiring physical contact means you have some choice as to who you communicate with, if you know your star trek , then you know how the alternative turns out for an unsuspecting Diana Troy Betazoid!

I don't fancy being violated in that manner - do you! , do you really like the idea of thought police either! cop.gif




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

We are now a read-only forum.
 
No new posts or registrations allowed.