Weka Software for Machine Learning and Data Mining
Weka is a set of software for machine learning and data mining
developed at the University of Waikato in New Zealand.
Weka is open source software issued under the GNU General Public License.
Getting Started with WEKA
From any workstation, open a secure shell to grendel:
chiron[1]% ssh grendel
From here, weka is found in /usr/local/weka. You merely have to
type "weka" to begin your session:
grendel[1]% weka
A java window will pop up, called the Weka GUI Chooser.
From there, select the Simple CLI (a command line version),
Explorer or Experimenter (java GUI based versions).
Weka can also be run from any PC that has SSH and X-window
clients (such as TeraTerm and X-Win32). Ensure that
TeraTerm has X Forwarding on by checking the box in
setup -> SSH Forwarding.
To close any window in Weka, press Alt-Q. You should
maximize the size of the GUI window to get the best
effects of the software's visualization.
Examples and Documentation
Sample data sets can be found on grendel in
/usr/local/weka/data/
There is an excellent tutorial in PDF format on how to use weka.
It can be found at
/usr/local/docs/WekaTutorial.pdf
There should not be any problems in getting the window to pop up on
your workstation by using SSH. If it is not working, you may need
to use an xhost command (see the man pages), or the display might be
incorrectly set.
grendel[1]% xhost +
grendel[1]% setenv DISPLAY $workstation:0
where $workstation = castor, pollux, etc.
e.g.,
grendel[1]% setenv DISPLAY castor:0
If problems persist, contact the system administrator by sending
mail to trouble@cs.uregina.ca.
