DM: RE: Bagging a classification Tree

"Bagging" is sort of a contraction of "bootstrap aggregating". Basically,
you generate a series of trees, each on a separate bootstrapped sample from
the original data. To classify new cases, you run them through all those
trees, and then combine the individual predictions across trees to produce
the bagged prediction.

You can find some useful information at Leo Breiman's ftp site (which you
can get to via his web page at http://www.stat.berkeley.edu/users/breiman/).

							--Clay

Clay Helberg                           http://www.execpc.com/~helberg/
SPSS Documentation                      chelberg@spss.com
Speaking only for myself....



 > -----Original Message-----
 > From: Mohamed A. Jeragh [mailto:maj@ecs.syr.edu]
 > Sent: Saturday, July 01, 2000 8:52 PM
 > To: datamine-l@nautilus-sys.com
 > Subject: DM: Bagging a classification Tree
 >
 >
 >
 > Can someone please post some links, or explain briefly what
 > is bagging a
 > a tree?
 > Thank you in advance,
 > --MJ
 >
 > 

• Prev by Date: Re: DM: Problem of Sample size
• Next by Date: DM: RE: Problem of Sample size
• Prev by thread: DM: RE: Problem of Sample size
• Next by thread: DM: Problem of Sample size
• Index(es):
  • Date
  • Thread