Orange Forum • View topic - Tree post-pruning and m_pruning

Tree post-pruning and m_pruning

A place to ask questions about methods in Orange and how they are used and other general support.

Tree post-pruning and m_pruning

Postby pulsoste » Wed Jul 18, 2012 9:48

Hi,
I'm building a classification tree with 1 million samples, on a two class problem. I'm wondering if using the m_pruning parameter = 2 as suggested before is a good number to do post pruning of the tree.

Does orange still implement the m estimates described in "Cestnik B. Estimating probabilities: A crucial task in machine learning. Proceedings of the Ninth European Conference on Artificial Intelligence", or has the implementation changed since 2005?

Thanks

Return to Questions & Support



cron