[R] computing misclassification table for tree objects

Marc R. Feldesman feldesmanm at pdx.edu
Sat Jan 5 05:18:52 CET 2002


At 04:36 PM 1/4/02, David Marimont wrote:
 >I have a classification tree that I computed via the tree function
 >(in the tree package).  I'd like to compute a misclassification table
 >(if that's the right term) on the data used to compute the tree.  That
 >is, I want to compute a table with the different classes (i.e.,levels of the
 >response factor) on the rows and the columns, and where entry [i,j] is
 >the number of times the tree classified an observation of class i as
 >one of class j.
 >

predict.tree(mytreeobject, type="class") will get you the predicted classes 
from the computed tree object (mytreeobject).  Suppose rawdata$realclass is 
the vector containing the original classifications used to generate 
mytreeobject.

Then, something like:

 >table(rawdata$realclass, predict(mytreeobject))  # or reversed if you like

should give you the contingency table (or confusion matrix) you're looking for.

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list