[om-list] Re: Method vs. Model, etc.

[Mark Butler]

Tom and other Packers wrote:

>     For me, there are two main objectives to writing good AI: (1) the model,
> (2) the method, i.e. the (1) statics and (2) dynamics of the language, i.e.
> (1) the data structure, and (2) the algorithm operating on that data; and
> it's almost arbitrary which one you start your work on, since they are both
> so interdependent.  I think Lee wants to see more of the algorithm part.

While there can be many methods, there can only be One Model in any sort of
unified system because every method has to operate on a form of data that
corresponds to a common model to avoid the NxN format conversion problem. 
 
>     I just discovered Support Vector Machine theory, which is very
> interesting, and similar to what I wanted to develop.  There are so many
> ideas out there, and they all seem to be overly concerned with method, above
> model, but I think we can learn something from them.

Any (web) references?

>     Here is my number-one desire right now, concerning our project:  How do
> we get data to test our method (our inferences) on, and how do we compare
> our results to the previously created systems so we know what value our
> system has, personal-use-wise, and market-wise?  I think we need to find
> data that other people have tested their programs on, and make similar
> tests.  Any thoughts?  Any known resources?

Well, I can get a lot of semi-questionable genealogical data.  Alternatively,
you could attempt natural language interpretation for simple sentences
extracted from a group of topically related web pages.  You could ask another
project for a list of common sense English sentences.

The Census department has lots of nice resources - the TIGER database has no
end of interesting social attributes if you want to classify localities.

If you drift off our current course feature extraction and classification from
sound samples, images, maps, etc, could all be done with easily obtainable
data.  Neural networks are generally much more suited to this type of
application - I can't imagine how you would make one that operated on a large
semantic database. 
 
- Mark