ANU/NICTA has several very active researchers working on SVMs; Many other
places around Australia also have good research programs related to these
topics.
Typically, for the purposes of classification, text files are encoded as
TF.IDF vectors. A simple Google search should turn up tons of material about
this encoding.
Balaji
----------------------------------------------------------------
Phone (home): 919-383-2069 | Off : 919-660-5233
Email : [log in to unmask] | http://www.duke.edu/~balaji/
----------------------------------------------------------------
----- Original Message -----
From: "Aneesha" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Friday, August 27, 2004 6:28 PM
Subject: Encoding Text for Classification
> Hi
>
> Thanks to everybody who sent links for papers and books that would help me
> learn to understand svms.
>
> I'm a little bit stuck on how I could encode text that I wish to classify
> to
> a dat file (just like in the examle I downloaded with svmLight) ie How do
> i
> calculate feature values like 0.43 and 0.12 eg
>
> -1 1:0.43 3:0.12 9284:0.2 # abcdef
>
> I have ordered a book on text classification using svms, but it will only
> arrive in a bout a month.
>
> Does anybody know of any intro courses/seminars to svms offered in
> Australia?
> Anybody from Australia and actively conducting research into SVMs?
>
> Thanks
>
> Aneesha
>
>
|