L. Kuncheva and C. Whitaker. Measures of Diversity in Classifier Ensembles and Their Relationship with the Ensemble Accuracy. Machine Learning, (51):181--207, 2003. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Constructing Diverse Classifier Ensembles using Artificial.. - Melville, Mooney (1 citation) (Correct)
....In an ensemble, the combination of the output of several classifiers is only useful if they disagree on some inputs [Krogh and Vedelsby, 1995] We refer to the measure of disagreement as the diversity of the ensemble. There have been several methods proposed to measure ensemble diversity [Kuncheva and Whitaker, 2002] usually dependent on the measure of accuracy. For regression, where the mean squared error is commonly used to measure accuracy, variance can be used as a measure of diversity. So the diversity of the i classifier on example x can be defined as: d i (x) C i (x) C (x) 2 where C i ....
L. Kuncheva and C. Whitaker. Measures of diversity in classifier ensembles and their relationship with the ensemble accuracy. submitted, 2002.
Re-designing Cost-Sensitive Decision Tree Learning - Estruch, Ferri.. (Correct)
....such as that one with the smallest size or with the lowest expected error. In order to do this, the hypotheses are compared with the combined hypothesis, which is used as an oracle. We have studied several measures of similarity which are based on metrics defined for a pair of hypothesis [12]. The metrics use a matrix that summarises which is the class predicted by a single hypothesis and the one predicted by the combined hypothesis. More formally, given two classifiers ha and hb, and an unlabelled dataset with n examples with C classes, we can construct a C x C contingency or ....
....cannot reserve part of the training set for this, or it could be counterproductive. The idea is then to use all the training set for constructing the hypotheses and a random dataset for selecting one of them. For this reason, we have considered the following measures which extend the metrics in [12] for dealing with classification problems which involve more than two classes and for using unlabelled datasets for the estimation of similarities: 0 measure: It is just based on the idea of reckoning the probability that both classifiers agree: Its value is between 0 and 1. An inverse measure, ....
[Article contains additional citation context not shown here]
Ludmila I. Kuncheva and Christopher J. Whitaker. Measures of diversity in classifier ensembles and their relationship with the ensemble accuracy. Submitted to Machine Learning, 2002.
Accuracy and Diversity in Ensembles of Text Categorisers - Adeva, Beresi (2005) (Correct)
No context found.
L. Kuncheva and C. Whitaker. Measures of Diversity in Classifier Ensembles and Their Relationship with the Ensemble Accuracy. Machine Learning, (51):181--207, 2003.
Accuracy and Diversity in Ensembles of Text Categorisers - Adeva, Beresi, Calvo (2005) (Correct)
No context found.
L. Kuncheva and C. Whitaker. Measures of Diversity in Classifier Ensembles and Their Relationship with the Ensemble Accuracy. Machine Learning, (51):181--207, 2003.
Set Analysis of Coincident Errors and Its Applications for.. - Ruta, Gabrys (Correct)
No context found.
Kuncheva L.I., C.J. Whitaker. Measures of diversity in classifier ensembles and their relationship with the ensemble accuracy, submitted.
Diversity in Ensemble Feature Selection - Tsymbal, Pechenizkiy, Cunningham (2003) (Correct)
No context found.
L.I. Kuncheva, C.J. Whitaker, Measures of diversity in classifier ensembles and their relationship with the ensemble accuracy, Machine Learning 51 (2) (2003) 181-207.
Constructing Diverse Classifier Ensembles using Artificial.. - Melville, Mooney (2003) (4 citations) (Correct)
No context found.
L. Kuncheva and C. Whitaker. Measures of diversity in classifier ensembles and their relationship with the ensemble accuracy. submitted, 2002.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC