Diversity measures for multiple classifier system analysis and design

  title={Diversity measures for multiple classifier system analysis and design},
  author={Terry Windeatt},
  journal={Information Fusion},
In the context of Multiple Classifier Systems, diversity among base classifiers is known to be a necessary condition for improvement in ensemble performance. In this paper the ability of several pair-wise diversity measures to predict generalisation error is compared. A new pair-wise measure, which is computed between pairs of patterns rather than pairs of classifiers, is also proposed for two-class problems. It is shown experimentally that the proposed measure is well correlated with base… CONTINUE READING
Highly Cited
This paper has 152 citations. REVIEW CITATIONS


Publications citing this paper.
Showing 1-10 of 61 extracted citations

Artificial Intelligence: Theories and Applications

Lecture Notes in Computer Science • 2012
View 20 Excerpts
Highly Influenced

Using Accuracy and Diversity to Select Classifiers to Build Ensembles

The 2006 IEEE International Joint Conference on Neural Network Proceedings • 2006
View 9 Excerpts
Highly Influenced

Tri_training for remote sensing classification based on multi-scale homogeneity

2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) • 2016
View 1 Excerpt

152 Citations

Citations per Year
Semantic Scholar estimates that this publication has 152 citations based on the available data.

See our FAQ for additional information.


Publications referenced by this paper.
Showing 1-10 of 10 references

Boosting he Margin : a new explanation for the effectiveness of voting methods

Y. Freund, P. Bartlett
Proc . Iberian Conf . On Pattern Recognition and Image Analysis • 2003

An exp rimental study on diversity for bagging and boosting with linear classifiers

M. Skurichina, R. P. W. Duin
Information Fusion • 2002

Similar Papers

Loading similar papers…