Re: DM: How to Update Decision Tree Dynamically

[Tjen-Sien Lim]

This issue has been addressed in ITI (the paper was published in
Machine Learning last year). Check out my decision trees
webpage for the link:

   http://www.stat.wisc.edu/~limt/treeprogs.html

-- 
Tjen-Sien Lim                (608) 262-8181                        
Ph.D. candidate              limt@stat.wisc.edu                    
Dept. of Statistics          http://www.stat.wisc.edu/~limt
Univ. of Wisconsin-Madison
1210 West Dayton Street       
Madison, WI 53706


   Date: Sun, 04 Oct 1998 13:06:40 -0600
   From: fardin akbaryan <fakbarya@ucalgary.ca>
   Organization: The University of Calgary
   X-Mailer: Mozilla 4.05 [en] (WinNT; I)
   MIME-Version: 1.0
   Content-Type: text/plain; charset=us-ascii
   Content-Transfer-Encoding: 7bit
   Sender: owner-datamine-l@nautilus-sys.com
   Precedence: bulk


   Dear Subscribers;

       My research is about pattern recognition for dynamic systems. 
As
   part of  my approach, I am using decision tree, C4.5/ C5 proposed 
by
   Quinlan, as the classifier. As the system receives the new 
information(
   pattern), I need to update my decision tree adaptivly by the most 
recent
   data while using the past data. In another word, the tree is 
supposed to
   combine fresh data and old ones. Obviously, I can make a new tree 
from
   the scratch by new data and part of old data. But I am interested 
in
   updating the decision tree on line as the new data have received 
by the
   system. I may define it as online tree updating.
       It will be highly appreciated if you give me information or 
any idea
   how to deal with that. So far I have not found a source in which 
this
   problem has been addressed. Thanks in advance for your attention


       Regards

   --


    Fardin Akbaryan
    Chemical and Petroleum Eng. Department
    University of Calgary
    2500 University Drive NW
    Calgary T2N 1N4, Alberta, Canada
    Tel: (403) 220-7409
    Fax: (403) 284-4852