Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

How to create one final decision tree

Hello,
I need to create on final decision tree based on 25 different decision trees but I am not sure how to do that. They need to be constructed in a sequential manner where we update the weights of the training examples based on the prediction and the error rate of the previous decision. Right now I have the random forest operator where I have 25 different decision trees but what is it that I used to create one final decision tree?

Thank you

Image: https://us.v-cdn.net/6030995/uploads/editor/ka/osd22p5csgti.png

Find more posts tagged with

AI Studio

Weights

Decision Tree

Tutorials + Training

Random Forest

Accepted answers

All comments

BalazsBaranyRM

Hi!

The process you're describing is implemented in ensemble models like Bagging and Boosting, and in Gradient Boosted Trees.

However, these create *separate* trees. I don't see how one decision tree would be able to express all the different trees adequately and I don't know an algorithm implementing this.

Regards,
Balázs