Random tree uses Entropy and IG?

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

Random tree uses Entropy and IG?

This post has NOT been accepted by the mailing list yet.
Hello everybody!

I'm trying tu uso random tree with some db's. I've read how decision tree choose which attributes select for each node, and so the use of entropy and Information gain.

The question is the following: does the random tree uses this 2 elements(entropy and ig)?
I've doubt about it and I can't understand if the algorithm choose for every node a random attribute.

Can be that the random tree select a random range of attributes and then use the entropy and info gain to select which split??

Thank you so much

Regards, AT