Re: [Gnumed-devel] need assessment from fellow clinicians

[Karsten Hilbert]

> Another thing that might be useful here is hidden markov model training 
> ;  as the user corrects the tentative
> tree updates, a hmm is trained to make a better parse next time . The 
> problem here might be ,
> hmm are being used to classify only 2 categories in practical use for 
> spam nowadays, but can gnumed
> use it to classify multiple categories ?
Also, the snippet of text being classified may be too small.
We are dealing with a few words only here, not comparable to
a full-featured spam messages (those tiny spams often pass the
HMMs, too).

Karsten
-- 
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346