Influence of Resampling and Weighting on Diversity and Accuracy of Classifier Ensembles

Authors:
R. M. Valdovinos;J. S. Sánchez;E. Gasca
Affiliations:
Lab. Reconocimiento de Patrones, Instituto Tecnológico de Toluca, Av. Tecnológico s/n, 52140 Metepec, México;Dept. Llenguatges i Sistemes Informátics, Universitat Jaume I Av. Sos Baynat s/n, E-12071 Castelló de la Plana, Spain;Lab. Reconocimiento de Patrones, Instituto Tecnológico de Toluca, Av. Tecnológico s/n, 52140 Metepec, México
Venue:
IbPRIA '07 Proceedings of the 3rd Iberian conference on Pattern Recognition and Image Analysis, Part II
Year:
2007

Citing 8
Cited 0

The weighted majority algorithm

Information and Computation
Bagging predictors

Machine Learning
An Experimental Comparison of Three Methods for Constructing Ensembles of Decision Trees: Bagging, Boosting, and Randomization

Machine Learning
Generating classifier outputs of fixed accuracy and diversity

Pattern Recognition Letters
Measures of Diversity in Classifier Ensembles and Their Relationship with the Ensemble Accuracy

Machine Learning
Complexity of Classification Problems and Comparative Advantages of Combined Classifiers

MCS '00 Proceedings of the First International Workshop on Multiple Classifier Systems
A new ensemble diversity measure applied to thinning ensembles

MCS'03 Proceedings of the 4th international conference on Multiple classifier systems
Evaluation of diversity measures for binary classifier ensembles

MCS'05 Proceedings of the 6th international conference on Multiple Classifier Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Diversity in the decisions of a classifier ensemble appears as one of the main issues to take into account for its construction and operation. However, the potential relationship between diversity and accuracy, with respect to the resampling method and/or the classifier fusion technique has not been clearly proved. The present paper analyzes the influence of different resampling methods and dynamic weighting schemes on diversity and how this can affect to the accuracy of the classifier ensemble. This is specifically studied in the framework of the Nearest Neighbor classification algorithm.