A dynamic overproduce-and-choose strategy for the selection of classifier ensembles
- 1 October 2008
- journal article
- Published by Elsevier in Pattern Recognition
- Vol. 41 (10), 2993-3009
- https://doi.org/10.1016/j.patcog.2008.03.027
Abstract
No abstract availableKeywords
This publication has 21 references indexed in Scilit:
- From dynamic classifier selection to dynamic ensemble selectionPattern Recognition, 2007
- A study on the performances of dynamic classifier selection based on local accuracy estimationPattern Recognition, 2005
- Classifier selection for majority votingInformation Fusion, 2005
- Automatic recognition of handwritten numerical strings: a recognition and verification strategyPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Ensembling neural networks: Many could be better than allArtificial Intelligence, 2002
- Dynamic classifier selection based on multiple classifier behaviourPattern Recognition, 2001
- On combining classifiersIEEE Transactions on Pattern Analysis and Machine Intelligence, 1998
- The random subspace method for constructing decision forestsIEEE Transactions on Pattern Analysis and Machine Intelligence, 1998
- Combination of multiple classifiers using local accuracy estimatesIEEE Transactions on Pattern Analysis and Machine Intelligence, 1997
- Automated design of linear tree classifiersPattern Recognition, 1990