Gremlin wrote:
Quote:
patty <pattyNO (AT) SPAMicyberspace (DOT) net> wrote in
Can you give me an example of an 'organic search term' ?
Well I think it would be something like this:
ORGANIC query="What is an organic search term?">It is a TAG that Gremlin
invented to make searching for information online seem more
human.</ORGANIC
In the case of this particular tag since it is in this post, the AI would
likely respond with that answer to that question if it was already on a
similar topic to the rest of this post... and if this post was given a high
enough ranking, so that query did not compete with any other similar
querys. You could take an approach like ALICE does of course so the person
does not have to enter the exact query... but the closer they are to the
exact query the higher the chance it would respond with that organic search
term. I mean my idea is very vauge.. but the goal would be to keep the
concept as simple as possible, so that looks like a good start. |
Well i think we need to be clear about several things:
(1) The search engine assigns a set of posts to a query -- in other
words the query defines the context.
(2) Mechanisms are used to rank multiple posts in that context. Perhaps
things like:
(a) how many posts responded to the post
(b) how many posts referred to the post
(c) how many web pages referred to the post
(d) how many other terms salient to the context are contained in the post
Up to this point we have algorithms working with just the words of the
post uncorrupted by human hands.
Now let us suppose we get posters to enter their own manual meta data in
the posts themselves. Perhaps something like:
<submitMeta>
organic search
question: What is an organic search?
answer: <http://www.marketposition.com/articles/organic-search.htm>
organic search term
question: What is an organic search term?
answer: <http://groups.google.com/groups?q=%22organic+search+term%22>
answer: "It is a TAG that Gremlin invented to make searching for
information online seem more human."
semiotics subjective experience language culture environment
picture: <http://www.icyberspace.net/patty/diagrams/SEMIOT.JPG>
by: patty
</submitMeta>
What ranking and emphasis is given to the posters manual meta data? How
does it interact with the automated algorithms?
Also you might ask yourself why Google consistently gives better results
than Ask Jeeves <http://www.ask.com/> ?
I'm not saying anything, i'm just posing things to think about.
patty