Calculating classification accuracy manually using rpart output

Machine Learning and Modeling
1

Hi all,

I was wondering would anyone be able to manually calculate the classification accuracy of a model from just the following rpart output.

  1. root 18 10 Grigolino (0.1666667 0.4444444 0.3888889)
    2) Sugar>=19.29943 2 0 Barolo (1.0000000 0.0000000 0.0000000) *
    3) Sugar< 19.29943 16 8 Grigolino (0.0625000 0.5000000 0.4375000)
  2. Phentanol>=25.76422 10 3 Grigolino (0.1000000 0.7000000 0.2000000)
    12) Phentanol< 35.72422 6 0 Grigolino (0.0000000 1.0000000 0.0000000) *
    13) Phentanol>=35.72422 4 2 Barbera (0.2500000 0.2500000 0.5000000)
    26) Zinc< 1.42099 1 0 Barolo (1.0000000 0.0000000 0.0000000) *
    27) Zinc>=1.42099 3 1 Barbera (0.0000000 0.3333333 0.6666667) *
    7) Phentanol< 25.76422 6 1 Barbera (0.0000000 0.1666667 0.8333333)
    14) Sugar>=16.56456 1 0 Grigolino (0.0000000 1.0000000 0.0000000) *
    15) Sugar< 16.56456 5 0 Barbera (0.0000000 0.0000000 1.0000000) *

It is a simple made up dataset with just 18 observations but I do not think there is a way to group the incorrectly classified observations?

Thanks in advance
Marco

2

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.