Improving Boosting Methods by Generating Specific Training and Validation Sets
In previous researches it can been seen that Bagging, Boosting and Cross-Validation Committee can provide good performance separately. In this paper, Boosting methods are mixed with Bagging and Cross-Validation Committee in order to generate accurate ensembles and take benefit from all these alternatives. In this way, the networks are trained according to the boosting methods but the specific training and validation set are generated according to Bagging or Cross-Validation. The results show that the proposed methodologies BagBoosting and Cross-Validated Boosting outperform the original Boosting ensembles.
KeywordsEnsembles of ANN Specific sets Boosting alternatives
Unable to display preview. Download preview PDF.
- 1.Asuncion, A., Newman, D.: UCI machine learning repository, University of California, Irvine, School of Information and Computer Sciences (2007)Google Scholar
- 5.Freund, Y., Schapire, R.E.: Experiments with a new boosting algorithm. In: International Conference on Machine Learning, pp. 148–156 (1996)Google Scholar
- 6.Hernández-Espinosa, C., Torres-Sospedra, J., Fernández-Redondo, M.: New experiments on ensembles of multilayer feedforward for classification problems. In: Proceedings of IJCNN 2005, pp. 1120–1124 (2005)Google Scholar
- 9.Parmanto, B., Munro, P.W., Doyle, H.R.: Improving committee diagnosis with resampling techniques. In: Advances in Neural Information Processing Systems, pp. 882–888 (1996)Google Scholar
- 12.Torres-Sospedra, J.: Ensembles of Artificial Neural Networks: Analysis and Development of Design Methods. Ph.D Thesis, Universitat Jaume I (2011)Google Scholar