R help - Dec 2012 - e1071 SVM: Cross-validation error confusion matrix

Hi, 

I ran two svm models in R e1071 package: the first without cross-validation
and the second with 10-fold cross-validation. 

I used the following syntax: 

#Model 1: Without cross-validation: 
> svm.model <- svm(Response ~ ., data=data.df,
type="C-classification",
> kernel="linear", cost=1) 
> predict <- fitted(svm.model) 
> cm <- table(predict, data.df$Response) 
> cm 
#Model2: With 10-fold cross-validation: 
> svm.model2 <- svm(Response ~ ., data=data.df,
type="C-classification",
> kernel="linear", cost=1, cross=10) 
> predict2 <- fitted(svm.model2) 
> cm2 <- table(predict2, data.df$Response) 
> cm2 
However, when I compare cm and cm2, I notice that the confusion matrices are
identical although the accuracy of each model is diffent. What am I doing
wrong? 
  
Thanks for you help, 




-----
TO GET MORE DETAILS CLICK HERE  
--
View this message in context:
http://r.789695.n4.nabble.com/e1071-SVM-Cross-validation-error-confusion-matrix-tp4651652.html
Sent from the R help mailing list archive at Nabble.com.

What do you mean by accuracy?  Proportion classified correctly is not a good
index of accuracy if that's the problem.
Frank

rahul143 wrote
> Hi, 
> 
> I ran two svm models in R e1071 package: the first without
> cross-validation and the second with 10-fold cross-validation. 
> 
> I used the following syntax: 
> 
> #Model 1: Without cross-validation: 
>> svm.model <- svm(Response ~ ., data=data.df,
type="C-classification",
>> kernel="linear", cost=1) 
>> predict <- fitted(svm.model) 
>> cm <- table(predict, data.df$Response) 
>> cm 
> 
> #Model2: With 10-fold cross-validation: 
>> svm.model2 <- svm(Response ~ ., data=data.df,
type="C-classification",
>> kernel="linear", cost=1, cross=10) 
>> predict2 <- fitted(svm.model2) 
>> cm2 <- table(predict2, data.df$Response) 
>> cm2 
> 
> However, when I compare cm and cm2, I notice that the confusion matrices
> are identical although the accuracy of each model is diffent. What am I
> doing wrong? 
>   
> Thanks for you help,




-----
Frank Harrell
Department of Biostatistics, Vanderbilt University
--
View this message in context:
http://r.789695.n4.nabble.com/e1071-SVM-Cross-validation-error-confusion-matrix-tp4651652p4651692.html
Sent from the R help mailing list archive at Nabble.com.