STRUCTURAL SYNTHESIS OF HYBRID NEURAL NETWORKS ENSEMBLES

V. M. Sineglazov, O. I. Chumachenko, O. R. Bedukha

Abstract


It is considered the structural synthesis of hybrid neural networks ensembles. It is chosen the ensemble topology as parallel structure with united layer. It is developed a hybrid algorithm for the problem solution which includes some algorithms preliminary choice of classifiers(modules of neural networks-hybrid neural networks, which consist of Kohonen, basic neural networks and bi-directional associative memory), creation the bootstrap training samples for every classifier, training these classifiers, optimal choice of necessity ones, determination of layer union weight coefficients, ensemble pruning. For the solution of optimal choice classifiers it is used two criteria: accuracy and variety.


Keywords


Hybrid neural networks ensembles; classifier; bootstrap training sample

References


I. Partalas, G. Tsoumakas, and I. Vlahavas, “Focused ensemble selection: A diversity-based method for greedy ensemble selection,” іn Proceeding of the 18th European Conference on Artificial Intelligence, 2008, pp. 117–121.

Y. Zhang, S. Burer, and W. N. Street, “Ensemble pruning via semi-definite programming,” The Journal of Machine Learning Research. pp. 1315–1338, 2006.

Lu. Zhenyu, Wu Xindong, Zhu Xingquan, and Josh Bongard, “Ensamble Pruning via Individual Contribution Ordering,” Department of Computer Science University of Vermont, Burlington. NSW. 2007, pp. 635–745.

Gonzalo Martґınez-Muñoz, and Alberto Suárez, “Aggregation Ordering in Bagging,” in Proc. of the IASTED International Conference on Artificial Intelligence and Applications. Acta Press, 2004, pp. 258–263.

G. Martıґnez-Muñoz and A. Suárez, “Pruning in ordered bagging ensembles,” in Proc. of the 23rd International Conference on Machine Learning, 2006, pp. 609–616.

Anne M. P. Canuto, Marjory C. C. Abreu, Lucas de Melo Oliveira, João C. Xavier Jr., and Araken de M. Santos, “Investigating the influence of the choice of the ensemble members in accuracy and diversity of selection-based and fusion-based methods for ensembles,” Pattern Recognition Letters, 28, 2007, pp. 472–486.


Full Text: PDF

Refbacks

  • There are currently no refbacks.


Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.