The investigation of model selection criteria in artificial neural networks by the Taguchi method

Tortum, Ahmet; Yayla, Nadir; Çelik, Cafer; Gökdağ, Mahir

doi:10.1016/j.physa.2007.07.064

Cited by 31 publications

(16 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In most researches, only the number of neuron in the hidden layer is optimized to find the optimum output. Though, there are also some researches which use more factors to find the optimum output, such as researches proposed by Sukhtomya and Tannock [14], Tortum, Yayla, Çelik, and Gökdağ [15], Lasheras, Vilán, Nieto, and Díaz [6]. The factors are the transformation data, the number of the training data, the number of the neuron in the input layer, the number of the neuron in the hidden layer, and the activation function.…”

Section: Introductionmentioning

confidence: 99%

Design of Experiment to Optimize the Architecture of Wavelet Neural Network for Forecasting the Tourist Arrivals in Indonesia

Otok

Suhartono

Ulama

et al. 2011

Informatics Engineering and Information Science

View full text Add to dashboard Cite

Abstract. Wavelet Neural Network (WNN) is a method based on the combination of neural network and wavelet theories. The disadvantage of WNN is the lack of structured method to determine the optimum level of WNN factors, which are mostly set by trial and error. The factors affecting the performance of WNN are the level of MODWT decomposition, the wavelet family, the lag inputs, and the number of neurons in the hidden layer. This research presents the use of design of experiments for planning the possible combination of factor levels in order to get the best WNN. The number of tourist arrivals in Indonesia via Soekarno-Hatta airport in Jakarta and via Ngurah Rai airport in Bali is used as case study. The result shows that design of experiments is a practical approach to determine the best combination of WNN factor level. The best WNN for data in Soekarno-Hatta airport is WNN with level 4 of MODWT decomposition, Daubechies wavelet, and 1 neuron in the hidden layer. Whereas, the best WNN for data in Ngurah Rai airport is WNN with MODWT decomposition level 3 and using input proposed by Renaud, Starck, and Murtagh [11] and seasonal lag input addition.

show abstract

Section: Introductionmentioning

confidence: 99%

Design of Experiment to Optimize the Architecture of Wavelet Neural Network for Forecasting the Tourist Arrivals in Indonesia

Otok

Suhartono

Ulama

et al. 2011

Informatics Engineering and Information Science

View full text Add to dashboard Cite

show abstract

“…The underlying idea of the information-based criteria is to identify an optimal trade-off between an unbiased approximation of the model and the complexity of the model. In [25,26], the authors use AIC in the neural network model selection to determine the optimal parameters of the ANN model. In this study, the comparison of ANN models is conducted using AIC.…”

Section: Akaike Information Criterion (Aic)mentioning

confidence: 99%

Application of Entropy-Based Attribute Reduction and an Artificial Neural Network in Medicine: A Case Study of Estimating Medical Care Costs Associated with Myocardial Infarction

Nie

Wang

2014

Entropy

View full text Add to dashboard Cite

Abstract:In medicine, artificial neural networks (ANN) have been extensively applied in many fields to model the nonlinear relationship of multivariate data. Due to the difficulty of selecting input variables, attribute reduction techniques were widely used to reduce data to get a smaller set of attributes. However, to compute reductions from heterogeneous data, a discretizing algorithm was often introduced in dimensionality reduction methods, which may cause information loss. In this study, we developed an integrated method for estimating the medical care costs, obtained from 798 cases, associated with myocardial infarction disease. The subset of attributes was selected as the input variables of ANN by using an entropy-based information measure, fuzzy information entropy, which can deal with both categorical attributes and numerical attributes without discretization. Then, we applied a correction for the Akaike information criterion ( ) to compare the networks. The results revealed that fuzzy information entropy was capable of selecting input variables from heterogeneous data for ANN, and the proposed procedure of this study provided a reasonable estimation of medical care costs, which can be adopted in other fields of medical science.

show abstract

“…Khaw et al (1995) used the Taguchi method as well as two simulated data collections to determine the effective parameters of ANN, which caused to increase the velocity and convergence of the Back Propagation (BP) algorithm. Also, other similar researches have been proposed in this field, such as the studies of Kim and Yum (2004), Sukthomya and Tannock (2005), Tortum et al (2007), Packianather et al (2000), and Peterson et al (1995). Bashiri and Geranmayeh (2011) proposed a method for tuning the parameters of the artificial neural network based on CCD and genetic algorithm.…”

Section: Tune-effective Parameters Of Ann For Estimation Of Responsesmentioning

confidence: 99%

A neuro-data envelopment analysis approach for optimization of uncorrelated multiple response problems with smaller the better type controllable factors

2013

View full text Add to dashboard Cite

In this paper, a new method is proposed to optimize a multi-response optimization problem based on the Taguchi method for the processes where controllable factors are the smaller-the-better (STB)-type variables and the analyzer desires to find an optimal solution with smaller amount of controllable factors. In such processes, the overall output quality of the product should be maximized while the usage of the process inputs, the controllable factors, should be minimized. Since all possible combinations of factors' levels, are not considered in the Taguchi method, the response values of the possible unpracticed treatments are estimated using the artificial neural network (ANN). The neural network is tuned by the central composite design (CCD) and the genetic algorithm (GA). Then data envelopment analysis (DEA) is applied for determining the efficiency of each treatment. Although the important issue for implementation of DEA is its philosophy, which is maximization of outputs versus minimization of inputs, this important issue has been neglected in previous similar studies in multi-response problems. Finally, the most efficient treatment is determined using the maximin weight model approach. The performance of the proposed method is verified in a plastic molding process. Moreover a sensitivity analysis has been done by an efficiency estimator neural network. The results show efficiency of the proposed approach.

show abstract

The investigation of model selection criteria in artificial neural networks by the Taguchi method

Cited by 31 publications

References 16 publications

Design of Experiment to Optimize the Architecture of Wavelet Neural Network for Forecasting the Tourist Arrivals in Indonesia

Design of Experiment to Optimize the Architecture of Wavelet Neural Network for Forecasting the Tourist Arrivals in Indonesia

Application of Entropy-Based Attribute Reduction and an Artificial Neural Network in Medicine: A Case Study of Estimating Medical Care Costs Associated with Myocardial Infarction

A neuro-data envelopment analysis approach for optimization of uncorrelated multiple response problems with smaller the better type controllable factors

Contact Info

Product

Resources

About