Google Advertising Professionals Promotional Credits | Main | Sandboxing Your Competitor

Yahoo!'s Concept Network & SuperUnits

Bill Slawski started a thread over at Cre8asite Forums named Yahoo! Superunits: of signatures and co-occurence. In that thread, he discusses a new patent Yahoo! Search released (on April 14th) named Systems and methods for search processing using superunits.

Here is the patent's abstract:

In a search processing system, a concept network is generated from a set of queries by parsing the queries into units and defining various relationships between the units based in part on patterns of units that appear together in queries. Units in the concept network that have some similar characteristic(s) are grouped into superunits. For each superunit, there is a corresponding signature that defines the similar characteristic of the group. A query is processed by identifying constituent units, determining the superunit membership of some or all of the constituent units, and using that information to formulate a response to the query.

Bill tells us to look for "some new vocabulary words - a concept network, a unit, a superunit, and signatures." I briefly skimmed it, but the whole concept network seems very interesting. "A concept network is generated from a set of queries by parsing the queries into units and defining various relationships between the units, e.g., based on patterns of units that appear together in queries."



Like The Story? Vote For It On Yahoo Buzz! Or On Sphinn!

posted rustybrick in Yahoo! Search Optimization at April 21, 2005 9:21 AM Comments (5)

Comments

You said, Here is the patent's abstract - and boy, is that abstract. I need much more coffee in me this morning to make sense out of that paragraph. In fact, it sounds more like something you would see being produced by one of those auto-spam generators, where it uses the keyword phrases chosen in a random meaningless way - but in a way that fools the engines into thinking it makes sense. :)

 

No one can fool Yahoo!

 

"Concept network, a unit, a superunit, and signatures" are not new things. If you read between lines, co-occurrence theory of set units and degree of memberships can be seen all over the patent.

Orion

 

My thoughts exactly, Orion. I expected to see the word "co-occurrence" somewhere within the patent after reading through part of it, and yet was surprised to see it, too.

These are not new things. But, they are concepts that people should start learning more about.

We're going to try to break that patent application down into a little more accessible english, Donna. Might take a day or more, but it's probably one for people to get a good understanding of.

 

That patent sounds very similar to the 'shard' or cluster orgnaisation of data by Google I saw in a recent presentaion.

Check out this presentation around the 45min mark.
http://www.search-engine-war.co.uk/2005/04/google_behind_t.html

 

Post a comment (Note: Can Take 120 Seconds For Your Comment To Show Up)

Do you want us to save your personal Information?

Premium Sponsors + advertise

To subscribe to the Search Engine Roundtable, click here