Pruning is a technique in machine learning that reduces the size of decision trees.
There can be pre-pruning and post-pruning.
Pre-pruning will use early stopping of the building of the tree.
Post-pruning will be done after the tree has first been created in full size and detail.
The goal of pruning is to reduce the complexity of the final classifier, hence increasing
generalisation and reducing the chances on
overfitting.
