Re: from word list to bayesian node... How do I organize this.

No one is responding and I'm not sure why.  I'd appreciate if anyone
knew where to look for an answer, be it a book or a website.  Even if
don't know the answer and have any input, I would appreciate it.

Thanks.


[EMAIL PROTECTED] (Jeff) wrote in message news:<[EMAIL PROTECTED]>...
> I think I understand bayasian theory ok.  I understand that if I
> choose a discreet value for a node then it chages the probabilities of
> the values in other nodes.  I am working on something that is similar
> to spam filtering, except that the data I'm looking to predict won't
> be binary (spam vs no spam).  I can't just give particular words a
> high probability of predicting "true".  For example, I need "chest"
> and "pain" to predict "chest pain".  I have made networks like this
> before, the problem is that the nature of the data I am supplying is
> different.  A list of words (which is what makes it similar to spam
> detection) is different from an organized table of data.
> 
> So, how do I set up and train a network to do this.  Do I set up a
> node for each word in the string (1st word, 2nd word, etc) and train
> it on all permutations of the word list? (I think this might over
> train it to the longer ones.) Do I not understand something about the
> theory?
> 
> Thanks in advance.
>

[ comp.ai is moderated.  To submit, just post and be patient, or if ]
[ that fails mail your article to <[EMAIL PROTECTED]>, and ]
[ ask your news administrator to fix the problems with your system. ]