A new criteria for comparing neural networks and Bayesian classifier

Othman, Ibtissem Ben; Ghorbel, Faouzi

doi:10.1109/iccat.2013.6522025

Cited by 4 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to compare the neural and statistical classifiers, most of researchers try to compare their prediction accuracy while forgetting the NNs instability criterion. In (Othman, 2013) and (Othman, 2014), we have proven the instability of network classifier results compared to the statistical methods. The stability evaluation is based on estimating the error rate probability density function (pdf) of each classifier.…”

Section: Introductionmentioning

confidence: 96%

Stability Evaluation of Combined Neural Networks

Othman

Ghorbel

2014

Proceedings of the International Conference on Neural Computation Theory and Applications

Self Cite

View full text Add to dashboard Cite

In the industrial field, the artificial neural network classifiers are currently used and they are generally integrated of technologic systems which need efficient classifier. However, the lack of control over its mathematical formulation explains the instability of its classification results. In order to improve the prediction accuracy, most of researchers refer to the classifiers combination approach. This paper tries to illustrate the capability of an example of combined neural networks to improve the stability criterion of the single neural classifier. The stability comparison is performed by the error rate probability densities function estimated by a new variant of the kernel-diffeomorphism semi-bounded Plug-in algorithm.

show abstract

Section: Introductionmentioning

confidence: 96%

Stability Evaluation of Combined Neural Networks

Othman

Ghorbel

2014

Proceedings of the International Conference on Neural Computation Theory and Applications

Self Cite

View full text Add to dashboard Cite

show abstract

“…The ANN surpasses the conventional Bayesian approach and other iterative techniques in pattern classification and segmentation because of the flexibility and suitability of the ANN technique for nonlinearity and dimensionality reduction [14][15][16][17]. The ANN, either independently or combined with the Bayesian rule, provides better results for image analysis [15][16][17][18][19][20]. Si and He [17] have applied an ANN technique to estimate the parameters of the autoregressive model and have reported that the ANN-based parameter estimate yields better results than the maximum likelihood, Bayes, and iterative methods.…”

Section: Introductionmentioning

confidence: 99%