[R] how to get confidential interval for classification accuracy using 10-fold validation in svm package

larry88 lglu88 at yahoo.com
Tue Jul 15 21:34:09 CEST 2008


I am just starting to use R language, and jumped into using svm (e1071) model
to do classification analysis using svm. Meanwhile, I used 10-fold cross
validation to evaluate initial classification accuracy. Now I am going to
run, for example, 100 times 10-fold cross validation to determine the
distribution of classification accuracy (confidential interval of
classificaiton accuracy). How do I program it to get these results? Thanks
in advance.
-- 
View this message in context: http://www.nabble.com/how-to-get-confidential-interval-for-classification-accuracy-using-10-fold-validation-in-svm-package-tp18473138p18473138.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list