https://www.displayr.com/machine-learning-pruning-decision-trees/

University of Michigan - Ann Arbor

Pruning decision trees is the process of removing certain branches which are not beneficial towards the model’s goal. Decision trees frequently over-fit, which leads to an excess of branches. As a result, a decision tree performs extremely well on the training set, but is over-fitting leading to poor test set performance. To prevent this, individuals can ‘prune’ the decision tree leading to fewer branches preventing over-fitting. Types of pruning include minimum error, smallest tree, and early stopping.

Learn Before

Related