New Chinese Search Engine - Accoona

Update 2004/12/07 - Accoona now live.

[Woz]

In this article [chinadaily.com.cn], China Daily Information Company and China Communications Corp have announced a partnership to develop a business search engine "utilizing a new and proprietary Artificial Intelligence Technology". China Communications Corp has filed a patent for the technology.

Although the reported database includes business worldwide, directly sourced rather than being spidered, it is unclear at this stage whether the engine will target purely Chinese Business or have a wider reach.

Onya
Woz

[gethan]

Spotted them on Alexa a few days ago - but didn't fully investigate - traffic rise of 16,000%

Saw it was a search engine and figured - that'll be the SEO's checking out their rankings then :)

So it's a Chinese search engine huh? - anyone checked out how the Chinese SERPs look? (someone must be fluent here)

[mahlon]

the serps are so bad it's comic.....

[bd4rr]

I did some quick test. A four-word keyword in English get 3 relative sponsor's link, most of the regular results are off topic. The same keyword gets totally different result from it's chinese site, #1 is a VERY spammy dating site. No, the keyword has nothing to do with that.

Another 2-word keyword in Chinese (similiar to the english one) get almost nothing useful.

Not useful at this time, in my opinion.

Ken

[tomkee]

More hype than substance.
Results so bad they should call it 'Artificial Stupidity' instead of 'Artificial Intelligence'

[moishe]

Well I am ranking quite nicely so I guess I can't complain. Anyone figure out who is supplying the PPC ads at the top of their serps?

[moishe]

DUH, just looked at properties, it is Overture.

Are they promoting this thing in the US at all?

[adfree]

What's their spider, didn't find any info. Look for their spider as a search on their own engine and you roll over laughing.

[Res_Ipsa]

I've been looking for spider identification information.

The "Submit Your Site" page states: We add and update new sites to our index each time we crawl the web,..."

And Accoona is described in the press as a crawler search engine.

Anyone?

[moishe]

no luck figuring out the spider here, anyone?

[bill]

Not yet. I even mailed them and asked as well, but there's been no response yet.

[GaryK]

Did you notice the form they use to submit a site? It uses a mailto: as the form's action property. How terribly unprofessional. I am still awaiting a crawl and a reply.

[Res_Ipsa]

From a July 8, 2004 article: "By comparison with general search systems, Acoona doesn't depend on Internet search systems known as spiders to 'crawl' across Web sites to locate fresh information. Instead it has signed commercial deals with private information providers." [usatoday.com...]

[bill]

...the plot thickens.

So, are they a crawler or not? That's a very interesting find Res_Ipsa. I don't recall seeing that in any of the other articles. Just where do they get their data from?

First there was the confusion about their name...just how do you spell that? Now we've got contradictory messages about whether they're a crawler or not. Not a great start...I know what you all think of the SERPs. ;)

[Res_Ipsa]

Doesn't appear to be a crawler. No spider identified and no guidelines for site inclusion. The web content is likely Yahoo's.

[adfree]

Now in the middle of the Google update: can I see that Accoona crap again...?

[Res_Ipsa]

Just curious about why the thread is under "Asia and Pacific Region?" The site is in English, the servers and the companies are based in US.

[bill]

You may want to re-read the thread again, particularly the article links.

One of the largest partners in this venture is the China Daily Information Company [chinadaily.com.cn] which is supported by the Chinese government. Although they do have an English interface as well, this engine was originally noticed in the press for its China focus. Their Chinese language interface is here: [accoona.cn...]

[Powdork]

The properties for all the links in results are as such htt*://www20.overture.com/d/sr/?xargs=15KPjg with the number after the www being the only variable. That should suggest where the content comes from.