Encyclopedia of Machine Learning

2010 Edition
| Editors: Claude Sammut, Geoffrey I. Webb


  • Johannes Fürnkranz
Reference work entry
DOI: https://doi.org/10.1007/978-0-387-30164-8_681


Pruning describes the idea of avoiding  Overfitting by simplifying a learned concept, typically after the actual induction phase. The word originates from  Decision Tree learning, where the idea of improving the decision tree by cutting some of its branches is related to the concept of pruning in gardening.

One can distinguish between  Pre-Pruning, where pruning decisions are taken during the learning process, and  Post-Pruning, where pruning occurs in a separate phase after the learning process. Pruning techniques are particularly important for state-of-the-art decision tree and  Rule Learning algorithms.

The key idea of pruning is essentially the same as  Regularization in statistical learning, with the key difference that regularization incorporates a complexity penalty directly into the learning heuristic, whereas pruning uses a separate pruning criterion or pruning algorithm.

Cross References

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  • Johannes Fürnkranz
    • 1
  1. 1.TU DarmstadtFachbereich InformatikDarmstadtGermany