Pruning (also called as inverse splitting) is a term in mathematics and informatics which allows to cut parts of a decision tree. Pruning is done by taking two leaves that have a common parent and merging them if it helps to improve the classification accuracy of decision tree. Pruned parts of the tree are no longer considered because the algorithm knows based on already collected data (e.g. through sampling) that these subtrees do not contain the searched object. The decision tree is simplified by removing some decisions. Pruning increases generalization.

Alumni Liaison

Ph.D. on Applied Mathematics in Aug 2007. Involved on applications of image super-resolution to electron microscopy

Francisco Blanco-Silva