Perturbation method for deleting redundant inputs of perceptron networks

Żurada, Jacek M.; Malinowski, A.; Usui, Shiro

doi:10.1016/s0925-2312(96)00031-8

Cited by 144 publications

(76 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The influence of the inputs over the output k is measured by the mean square average sensitivities [19], S ik,avg , that represent an overall sensitivity of each input variable over the outputs. The input with the lesser value of sensitivity will be assumed to be unnecessary and will be remove from the input set.…”

Section: B Variables That Influence Seepage Phenomenonmentioning

confidence: 99%

Dam seepage analysis based on artificial neural networks: The hysteresis phenomenon

Santillán

Fraile-Ardanuy

Toledo

2013

The 2013 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

Section: B Variables That Influence Seepage Phenomenonmentioning

confidence: 99%

Dam seepage analysis based on artificial neural networks: The hysteresis phenomenon

Santillán

Fraile-Ardanuy

Toledo

2013

The 2013 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

“…Methods described in [8] and [41] use genetic algorithms to select the relevant feature subsets. Methods described in [3], [13], [43]- [45], and [49] and a variety of others use neural networks for feature selection. Feature selection has also been attempted using fuzzy and neurofuzzy techniques [12], [42].…”

mentioning

confidence: 99%

Selecting Useful Groups of Features in a Connectionist Framework

Chakraborty

Pal

2008

IEEE Trans. Neural Netw.

View full text Add to dashboard Cite

Abstract-Suppose for a given classification or function approximation (FA) problem data are collected using sensors. From the output of the th sensor, features are extracted, thereby generating = =1 features, so for the task we have as input data along with their corresponding outputs or class labels . Here, we propose two connectionist schemes that can simultaneously select the useful sensors and learn the relation between and . One scheme is based on the radial basis function (RBF) network and the other uses the multilayered perceptron (MLP) network. Both schemes are shown to possess the universal approximation property. Simulations show that the methods can detect the bad/derogatory groups of features online and can eliminate the effect of these bad features while doing the FA or classification task.

show abstract

“…Two variable selection techniques were used to reduce redundant or irrelevant information gathered during data set preparation: based on the methodology proposed by Żurada et al 19 with further modification by Mendyk and Jachowicz, 20 which uses ANNs trained with the back propagation algorithm as the modeling tools; and feature ranking created by the fscaret package of the R environment (The R Foundation for Statistical Computing, Vienna, Austria). 21,22 The main parameters of the applied techniques are listed in Table 1.…”

Section: Feature Selectionmentioning

confidence: 99%

Heuristic modeling of macromolecule release from PLGA microspheres

Szlęk¹,

Pacławski²,

Lau³

et al. 2013

IJN

View full text Add to dashboard Cite

Dissolution of protein macromolecules from poly(lactic-co-glycolic acid) (PLGA) particles is a complex process and still not fully understood. As such, there are difficulties in obtaining a predictive model that could be of fundamental significance in design, development, and optimization for medical applications and toxicity evaluation of PLGA-based multiparticulate dosage form. In the present study, two models with comparable goodness of fit were proposed for the prediction of the macromolecule dissolution profile from PLGA micro- and nanoparticles. In both cases, heuristic techniques, such as artificial neural networks (ANNs), feature selection, and genetic programming were employed. Feature selection provided by fscaret package and sensitivity analysis performed by ANNs reduced the original input vector from a total of 300 input variables to 21, 17, 16, and eleven; to achieve a better insight into generalization error, two cut-off points for every method was proposed. The best ANNs model results were obtained by monotone multi-layer perceptron neural network (MON-MLP) networks with a root-mean-square error (RMSE) of 15.4, and the input vector consisted of eleven inputs. The complicated classical equation derived from a database consisting of 17 inputs was able to yield a better generalization error (RMSE) of 14.3. The equation was characterized by four parameters, thus feasible (applicable) to standard nonlinear regression techniques. Heuristic modeling led to the ANN model describing macromolecules release profiles from PLGA microspheres with good predictive efficiency. Moreover genetic programming technique resulted in classical equation with comparable predictability to the ANN model.

show abstract

Perturbation method for deleting redundant inputs of perceptron networks

Cited by 144 publications

References 6 publications

Dam seepage analysis based on artificial neural networks: The hysteresis phenomenon

Dam seepage analysis based on artificial neural networks: The hysteresis phenomenon

Selecting Useful Groups of Features in a Connectionist Framework

Heuristic modeling of macromolecule release from PLGA microspheres

Contact Info

Product

Resources

About