Looking at Yahoo's Personalization Patent

Source Title:
Yahoo!'s Concept Network & SuperUnits
Story Text:

Bragadoccio, aka Bill Slawski points out a new Yahoo Patent that outlines plans to base search results on the way people search. Following closely from news of Google's new Search History feature it's one more step down the personalization road...

From the patent:

What human beings think in terms of are natural concepts. For example, "hawaii" and "new york city" are vastly different queries in terms of length as measured by number of words but for a human being they share one important characteristic: they are each made up of one concept. In contrast, a person regards the query "new york city law enforcement" as fundamentally different because it is made up of two distinct concepts: "new york city" and "law enforcement.

Human beings also think in terms of logical relationships between concepts. For example, "law enforcement" and "police" are related concepts since the police are an important agency of law enforcement; a user who types in one of these concepts may be interested in sites related to the other concept even if those sites do not contain the particular word or phrase the user happened to type. As a result of such thinking patterns, human beings by nature build queries by entering one or more natural concepts, not simply a variably long sequence of single words, and the query generally does not include all of the related concepts that the user might be aware of. Also, the user intent is not necessarily reflected in individual words of the query. For instance, "law enforcement" is one concept, while the separate words "law" and "enforcement" do not individually convey the same user intent as the words combined.

The cre8asite thread already thrashes out some of the implications, as does seroundtable author pheonix.

We talked about the public perception of SEO recently, and in that thread many compared SEO to PR (Public Relations). If personalization really does spell an end to easy SEO then SEO will surely move even closer to it's PR cousin im thinking...


New search engines, new SEO

In view of Michael Martinez' recent piece and a lot of other corrobative evidence this points to a) the search engines developing into data mining corps as their main (currently only secondary if not tertiary) focus; b) user behavior based ranking algorithms in the (mistaken) hope that these won't be as easy to manipulate by SEOs; c) more refined user demographics - see item a) above - which can be monetized via exponentially improved targeting as another enticement for advertisers.

Of course, new search engines will demand new SEO approaches - so, black hats: build your bot nets and human surfer simulator bots and generic vintage-to-be domains now or become roadkill in a year or a year and a half's time ...

As an aside: this has basically been predicted if not in as many words by Mike Grehan some 2 years ago. It's kinda funny that I was so tied up with other, non SEO related work last year I more or less missed out on jumping on the linking frenzy bandwaggon big time which, as many are now beginning to realize, was (and to a great extent still is) probably the greatest blunder the SEO industry committed in its short history, at least from a long term perspective. But then again, "long term SEO" is really a bit of an oxymoron, isn't it? :-)


I have a lot of problems with personalized searching. I think John Q. Public will have an even bigger one, because they don't understand it.

You search for something at home one day and click on the first result and don't book mark it. Try to to do the same search at the office the next day and get a different result, you would get pretty annoyed. If you tried the search at home a few days later and can't find it, because the kids and your spouse have searched and altered the profile, how annoying would it be.

Imagine if everytime you went to the supermarket it was a different layout based on what you shopped for last time. They may put the toilet paper on the end-cap because it's on sale, but as someone who does all the food shopping I can tell you people look/shop/take it from the aisle first because they know thats where it is.

I do agree that SE's are experimenting with this big time though, and it's going to change things. I just don't think it's a good idea.

I just don't think it's a good idea.

Neither do I - it will be defeated just like all other search algos and spam protection strategies were.
The probable difference being that, as usual, only the big boys will survive because of the enormous hardware and bandwidth and maintenance and personnel costs involved, not to mention R&D and the sheer brainpower required to keep up with what's coming.

And yes - it will probably be about as efficiently "personalized" as Mail Merge. You know, when you get these nice serial letters starting "Dear Mr High Street 55 ..."


Granted, I haven't read every single word of the application, but guess I have to question the use of "personalization" in this context.

I'm probably missing something, but I just see a different way of parsing a search request, not anything about using a searcher's history or anything along those lines. About the only "user behaviour" factor I can find is a la Direct Hit, with aggregated click through data.

One of the most interesting pieces to me as basing "seeds" on external sources:

Examples of external sources include a list of related terms created by an editor or editorial team (e.g., a list of popular singers or a list of auto manufacturers known to the team); an authoritative web site (e.g., a medical reference site that maintains a dictionary or other listing of diseases); or the like.

Not all algo dependent. That could be refreshing, or not.

(By the way, you really have to eyeball the [url=http://aiw2.uspto.gov/.aiw?Docid=20050080795&homeurl=http%3A%2F%2Fappft1.uspto.gov%2Fnetacgi%2Fnph-Parser%3FSect1%3DPTO2%2526Sect2%3DHITOFF%2526u%3D%25252Fnetahtml%25252FPTO%25252Fsearch-adv.html%2526r%3D29%2526f%3DG%2526l%3D50%2526d%3DPG01%2526p%3D1%2526S1%3D%25252820050414.PD.%252BAND%252B%252528search%252BAND%252Bengine%252529.BIS.%252529%2526OS%3Dpd%2F20050414%252Band%252Bspec%2F(search%252Band%252Bengine)%2526RS%3D(PD%2F20050414%252BAND%252BSPEC%2F(search%252BAND%252Bengine))&PageNum=&Rtype=&SectionNum=&idkey=5DFEF98B2128]classy figures[/url] they included in the application.)

Thank goodness!

Personalization will be great for the end user, and even better for those of us that strive to create high quality websites that encourage users to visit again and again. This is just the next basic step in the evolution of the web. Imagine if telivision or radio were as unfriendly as the web is. This is a giant step forward -- the only downside I see to it is this will only make the big players in search bigger, and make it that much harder for the smaller companies and start-ups to break into search.

While I don't share your enthusiasm

about Google or anyone else tracking my browsing habits and storing them till kingdom come to cut a buck, it's really of no import: personalized search will come, whether we like it (as you seem to) or not (as fossils like me do who still remember the age old concept of "privacy") ...

And yes - Mike also pointed out that even in SEO terms the "filthy rich" will get ever richer whereas the Great Unwashed won't stand a chance anymore.

Unless, of course, someone helps them level the playing field ...


I'm with you fanto. I may just go back to doing without search engines entirely. It's not that difficult.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.