Question

I’m comparing some classifiers. My procedure is to compute the confusion matrix with the table command, and then calculate the false positive and true positive rates from the table. The routine I wrote requires that the table be square. There should be an easy way to do it.

My setup:

cm <- table(classifiers$teacher[which(classifiers$problem == ‘problem27’)],
classifiers$srAve[which(classifiers$problem == ‘problem27’)])
cm

     1  2  3
  0 23  0  0
  1  2  4  0
  2  2 10  0
  3  0  1  0
  4  0  0  1
> missingNames <- as.numeric( rownames(cm)[ !(rownames(cm) %in% as.numeric(colnames(cm)))  ] )
> missingNames
[1] 0 4

And then the C like function I wrote to fix it:

padTable <- function( missingNames, cm ) {
    rowLength <- dim(cm)[1]
    for (i in missingNames) {
       zeroes <- rep(0,rowLength)
       cNames <- colnames(cm)
       after <- which ( (i < as.numeric(cNames)) )[1]
       before <- which ( (i > as.numeric(cNames)) )[1]
       if ( is.na(before) ) { #The very begining
          cm <- cbind(zeroes,cm)
          colnames(cm) <- c(i,cNames)
       } else {
          if (is.na(after)) { #The very end
              cm <- cbind(cm,zeroes)
              colnames(cm) <- c(cNames,i)
          } else { #somewhere in the middle
               print('ERROR CANNOT INSERT INTO MIDDLE YET.')
               cm = NULL
          }
       }
    } 
    return(cm)
}

So, there has to be some dreadfully simple way to make this work. Anytime I find myself writing C code in R, I know that I’m doing it wrong.

Thanks for any help.

EDIT: Sample data as requested:

> classifiers$teacher[which(classifiers$problem == 'problem27')]
[1] 0 0 1 2 2 2 0 0 0 0 0 2 0 0 2 0 4 3 0 0 2 2 0 2 0 0 2 2 1 0 0 2 1 0 1 2 0 0
[39] 0 1 0 0 1
> classifiers$srAve[which(classifiers$problem == 'problem27')]
[1] 1 1 2 2 2 2 1 1 1 1 1 1 1 1 1 1 3 2 1 1 2 2 1 2 1 1 2 2 1 1 1 2 2 1 2 2 1 1
[39] 1 2 1 1 1

Answer 1

You should simply be able to convert classifiers$teacher and classifiers$srAve to factors but I’m just guessing, since I don’t know what your data are like.

> x <- factor(sample(0:4,20,TRUE))
> y <- factor(sample(1:3,20,TRUE),levels=levels(x))
> z <- data.frame(x,y)
> table(z)
   y
x   0 1 2 3 4
  0 0 1 2 0 0
  1 0 1 0 1 0
  2 0 2 2 1 0
  3 0 3 3 2 0
  4 0 0 2 0 0
> z$y <- as.character(y)
> table(z)
   y
x   1 2 3
  0 1 2 0
  1 1 0 1
  2 2 2 1
  3 3 3 2
  4 0 2 0