PDA

View Full Version : Decision tree HELP, Please:)



darklightred
04-15-2007, 01:21 AM
Can anyone please help me undestand the procedure of Decision tree in Weka?

what are the steps?
and how does it evaluate the attributes?

thanks

mthmulders
04-17-2007, 05:51 PM
It depends on the algorithm being used to construct a tree. The most well-known (I guess) is ID3 (developped by Quinlan), so let's take ID3 as an example.

ID3 splits the data in two parts. The exact criterion is determined by examining the entropy of the two subsets. The split that results in the largest information gain (or decrease in entropy) is executed.
This procedure is then recursively applied on the subsets of the dataset.

Maybe my explanation is not perfect, but at least it might give you an idea of how things work.