Decision Tree vs ID3

tanthiamhuat

could anyone explain to me what is the difference when using either Decision Tree or ID3 in RapidMiner?
I try both Decision Tree and ID3 on a same DataSet_A, they produce different outputs.
Again, I try Decision Tree and ID3 on a same DataSet_B, they produce same outputs.

So what are the criteria that determines whether same or different outputs? what how do we know which to use?
I use Information Gain for both.

MartinLiebig

Hi,

if i remember it correctly the standard Rapidminer Decision Tree implements both ID3 and CHAID, depending on which criterion you use.
The big benefit of using RM Decision Tree (v6.3+) is, that the decision tree is running in parallel. The Weka models don't. So i woud prefer the RM one first place.

~Martin