Hypothesis Diversity in Ensemble Classification
The paper discusses the issue of hypothesis diversity in ensemble classifiers. The measures of diversity previously proposed in the literature are analyzed inside a unifying framework based on Monte Carlo stochastic algorithms. The paper shows that no measure is useful to predict ensemble performance, because all of them have only a very loose relation with the expected accuracy of the classifier.
KeywordsDiversity Measure Unify Framework Entropy Measure Machine Learning Research Neural Network Ensemble
Unable to display preview. Download preview PDF.
- 1.Dietterich, T.: Machine learning research: Four current directions (1997)Google Scholar
- 2.Esposito, R., Saitta, L.: A monte carlo analysis of ensemble classification. In: Greiner, R., Schuurmans, D. (eds.) Proceedings of the twenty-first International Conference on Machine Learning, Banff, Canada, July 2004, pp. 265–272. ACM Press, New York (2004)Google Scholar
- 3.Esposito, R., Saitta, L.: Experimental Comparison between Bagging and Monte Carlo Ensemble Classification. In: Raedt, L.D., Wrobel, S. (eds.) Proceedings of the twenty-second International Conference of Machine Learning, Bonn, Germany, August 2005, ACM Press, New York (2005)Google Scholar
- 5.Kuncheva, L.: That elusive diversity in classifier ensembles (2003)Google Scholar
- 7.Partridge, D., Krzanowski, W.: Distinct failure diversity in multiversion software (1997)Google Scholar