A new ensemble pruning method based on the margin concept

作者：Guo Li^*; Boukir Samia

来源：Traitement du Signal, 2010, 27(6): 491-514.

DOI：10.3166/TS.27.491-514

• multiple classification
• ensemble method
• ensemble pruning
• margin
• decision tree

Ensemble methods have been successfully used as a classification scheme. The reduction of the complexity of this popular learning paradigm motivated the appearance of ensemble pruning algorithms. This paper presents a new efficient ensemble pruning method which not. only highly reduces the complexity of ensemble methods but also performs better than the non-pruned version in terms of classification accuracy This algorithm consists in ordering all the base classifiers with respect to their entropy which exploits a new version of the margin of ensemble methods. Confrontation with both the naive approach of randomly pruning base classifiers and another ordered-based pruning algorithm turned out convincing in an extensive empirical analysis.

• 出版日期2010