Re: Select Announment

Jason Rennie <[email protected]>:

> And ifile doesn't do any feature selection (the only reason it throws out 
> infrequent words is to keep the database size down).

yet ifile is simple and works very well.  i'm always intrigued when
going thru my ~/.idata.  it's ASCII and contains all that's needed in
a simple format.  back when i used it to classify my email into 20+
categories, the correct classification was always in the top three.
with categories sufficiently distinct, especially with the simple
spam/non-spam pair, it's accuracy is amazing.

> Feature selection: choosing features for classification.  In documents, 
> features are normally words, but could be things like "the number of 
> exclamation points," "the number of ALL CAPS words" or "the number of 
> words that begin with the letter `x'" (good for identifying documents 
> about XWindows :)

thanks for making this sound simple.  i tried the first pages of your
aimemo2001 :)

clemens

------------------------ Yahoo! Groups Sponsor ---------------------~-->
Get 128 Bit SSL Encryption!
http://us.click.yahoo.com/JjlUgA/vN2EAA/kG8FAA/CefplB/TM
---------------------------------------------------------------------~->

To UNSUBSCRIBE, send an empty message to
[email protected]

ifile web site: http://www.ai.mit.edu/~jrennie/ifile/ 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/