Re: Best way to classify...

> Say, for example, that I have this piece of text which might belong to
> one of four classifications.
> Maybe it is spam, normal ham, ham about cats, ham about dogs. Can
> crm114 choose from multiple classifications at once?
> Or is it a binary decision sort of thing where I have to build a
> binary decision tree and consider spam vs ham and of the ham consider
> ham about cats vs ham about dogs?

I wouldn't say that it has to be a binary decision, but the choices  
you gave are all binary. So yeah, you'd make a tree, like

            ham             spam
            / \
   about cats  about dogs

But you could just as easily have 'ham' have ten categories or twenty.  
In pseudocode:

if classify(msg,['ham','spam']) == 'spam':
     file('spam')
else:
     category = classify(msg,['dogs', 'cats', 'bacon', 'cheese'])
     file(category)

Or at least, such a tree works well for my data. You might want to try  
it with yours before you take my word as gospel :)


------------------------------------------------------------------------------
This SF.net email is sponsored by:
High Quality Requirements in a Collaborative Environment.
Download a free trial of Rational Requirements Composer Now!
http://p.sf.net/sfu/www-ibm-com