topic 回复： What is the correct k-fold operation in a decision tree module? in Discussions

What is the correct k-fold operation in a decision tree module?

lala — Fri, 09 Jun 2023 18:04:29 GMT

If the k-fold operation of decision tree module is carried out manually,
Is k-fold selected after segmentation?Or is k-fold selected and then segmented?

What exactly does k-fold mean?

Thanks !

回复： What is the correct k-fold operation in a decision tree module?

lala — Fri, 05 Nov 2021 07:45:32 GMT

The second result in the figure is the result of running the script in the example

dt = Open( "$SAMPLE_DATA/Car Poll.jmp" );
obj = dt << Partition( Y( :country ), X( :sex, :marital status, :age, :type, :size ), Split Best( 2 ) );
obj << K Fold Crossvalidation( 5 );

The script should perform the decision tree operation first and then the K-fold operation.

回复： What is the correct k-fold operation in a decision tree module?

Mark_Bailey — Fri, 05 Nov 2021 16:48:54 GMT

The K-fold method of cross-validation in Partition is problematic. The command has been deprecated starting in JMP 16. It is still available by scripting, but the method is not recommended.

You should learn about the methods before you use them. K-fold is an economic approach to applying cross-validation with small data sets that might suffer training and validation if hold out sets are created.