Fuzzy neural network modeling for product development

A fairly general product development model is formulated and analyzed based on multiple attribute decision making with emphasis on the treatment of the linguistic and vague aspects by fuzzy logic and up-dating or learning by neural network. Due to the representative ability of fuzzy set theory and the learning or intelligent ability of neural network, the proposed approaches appear to be an effective tool for handling vague and not well-defined systems.     

Fuzzy models and relational equations

We consider a problem of system modelling expressed mathematically in terms of fuzzy sets, especially fuzzy relational equations. A general methodological scheme is proposed and, as a sequel, particularized in the form of a sequence of several steps, such as: (i) structure determination, (ii) parameter determination and (iii) model validation. Procedures for studying and detecting a data structure are given special attention. New schemes are proposed. Moreover, several examples indicating fields of application of fuzzy models are considered.     

Fuzzy relational equations with min-biimplication composition

This paper discusses fuzzy relational equations with min-biimplication composition where the biimplication is the biresiduation operation with respect to the ?ukasiewicz t-norm. It is shown that determining whether a finite system of fuzzy relational equations with min-biimplication composition has a solution is NP-complete. Moreover, a system of such equations can be fully characterized by a system of integer linear inequalities and consequently its solution set can be expressed in the terms of the minimal solutions of this system of integer linear inequalities.     

Fuzzy relational equations and the inverse problem

The inverse problem concerned with fuzzy relations is investigated. The conditions for the existence of a solution are shown and an analytical solution is given. A method for the improvement of the solution is proposed.     

Comparing SOM neural network with Fuzzy c-means,K-means and traditional hierarchical clustering algorithms

In this paper we present a comparison among some nonhierarchical and hierarchical clustering algorithms including SOM (Self-Organization Map) neural network and Fuzzy c-means methods. Data were simulated considering correlated and uncorrelated variables, nonoverlapping and overlapping clusters with and without outliers. A total of 2530 data sets were simulated. The results showed that Fuzzy c-means had a very good performance in all cases being very stable even in the presence of outliers and overlapping. All other clustering algorithms were very affected by the amount of overlapping and outliers. SOM neural network did not perform well in almost all cases being very affected by the number of variables and clusters. The traditional hierarchical clustering and K-means methods presented similar performance.     

Fuzzy relational clustering around medoids: A unified view

Medoid-based fuzzy clustering generates clusters of objects based on relational data, which records pairwise similarities or dissimilarities among objects. Compared with single-medoid based approaches, our recently proposed fuzzy clustering with multiple-weighted medoids has shown superior performance in clustering via experimental study. In this paper, we present a new version of fuzzy relational clustering in this family called fuzzy clustering with multi-medoids (FMMdd). Based on the new objective function of FMMdd, update equations can be derived more conveniently. Moreover, a unified view of FMMdd and two existing fuzzy relational approaches fuzzy c-medoids (FCMdd) and assignment-prototype (A-P) can be established, which allows us to conduct further analytical study to investigate the effectiveness and feasibility of the proposed approach as well as the limitations of existing ones. The robustness of FMMdd is also investigated. Our theoretical and numerical studies show that the proposed approach produces good quality of clusters with rich cluster-based information and it is less sensitive to noise.     

Fuzzy regression analysis using neural networks

In this paper, we propose simple but powerful methods for fuzzy regression analysis using neural networks. Since neural networks have high capability as an approximator of nonlinear mappings, the proposed methods can be applied to more complex systems than the existing LP based methods. First we propose learning algorithms of neural networks for determining a nonlinear interval model from the given input-output patterns. A nonlinear interval model whose outputs approximately include all the given patterns can be determined by two neural networks. Next we show two methods for deriving nonlinear fuzzy models from the interval model determined by the proposed algorithms. Nonlinear fuzzy models whose h-level sets approximately include all the given patterns can be derived. Last we show an application of the proposed methods to a real problem.     

Integration of particle swarm optimization-based fuzzy neural network and artificial neural network for supplier selection

This study is intended to develop an intelligent supplier decision support system which is able to consider both the quantitative and qualitative factors. It is composed of (1) the collection of quantitative data such as profit and productivity, (2) a particle swarm optimization (PSO)-based fuzzy neural network (FNN) to derive the rules for qualitative data, and (3) a decision integration model for integrating both the quantitative data and fuzzy knowledge decision to achieve the optimal decision. The results show that the decision support system developed in this study make more precise and favorable judgments in selecting suppliers after taking into account both qualitative and quantitative factors.     

Fuzzy polynomial regression with fuzzy neural networks

Recently, fuzzy linear regression is considered by Mosleh et al. [1]. In this paper, a novel hybrid method based on fuzzy neural network for approximate fuzzy coefficients (parameters) of fuzzy polynomial regression models with fuzzy output and crisp inputs, is presented. Here a neural network is considered as a part of a large field called neural computing or soft computing. Moreover, in order to find the approximate parameters, a simple algorithm from the cost function of the fuzzy neural network is proposed. Finally, we illustrate our approach by some numerical examples.     

Efficiency decomposition in network data envelopment analysis: A relational model

Traditional studies in data envelopment analysis (DEA) view systems as a whole when measuring the efficiency, ignoring the operation of individual processes within a system. This paper builds a relational network DEA model, taking into account the interrelationship of the processes within the system, to measure the efficiency of the system and those of the processes at the same time. The system efficiency thus measured more properly represents the aggregate performance of the component processes. By introducing dummy processes, the original network system can be transformed into a series system where each stage in the series is of a parallel structure. Based on these series and parallel structures, the efficiency of the system is decomposed into the product of the efficiencies of the stages in the series and the inefficiency slack of each stage into the sum of the inefficiency slacks of its component processes connected in parallel. With efficiency decomposition, the process which causes the inefficient operation of the system can be identified for future improvement. An example of the non-life insurance industry in Taiwan illustrates the whole idea.     

Fuzzy adaptive network in presidential elections

Several scientific forecasting models for presidential elections have been suggested. However, most of these models are based on traditional statistics approaches. Since the system is linguistic, vague, and dynamic in nature, the traditional rigorous mathematical approaches are inappropriate for the modeling of this kind of humanistic system. This paper presents a combined neural fuzzy approach, namely a fuzzy adaptive network, to model and forecast the problem of a presidential election. The fuzzy adaptive network, which is ideally suited for the modeling of vaguely defined humanistic systems, combines the advantages of the representation ability of fuzzy sets and the learning ability of a neural network. To illustrate the approach, experiments were carried out by first formulating the problem, then training the network, and, finally, predicting the election results based on the trained network. The experimental results show that a fuzzy adaptive network is an ideal approach for the modeling and forecasting of national presidential elections.     

Fuzzy network technique for technological forecasting

The stochastic network technique is known to be a powerful tool carrying out a technological forecast of complex systems. A network dealt with is characterized by a tetrad of essential elements: logical nodes with some inputs and outputs, probabilistics activity branches, feedback loops, and multiple sources and sinks. A set of network parameters is defined for each element and their values are estimated for practical analysis of the network. In the case where the system to be treated is very large and/or complex, it cannot always be represented by a definite network and therefore forecasted values of parameters are inevitably indefinite themselves. A conventional probabilistic approach is sometimes inadequate in such a case. In the light of these facts, the paper proposes a fuzzy network technique, in which among activity branches emanating from a node, a branch to be undertaken once the node is realized belongs to a fuzzy set; and the time required to complete an activity branch belongs to a fuzzy set. Operations of maximum and minimum for sum and product of fuzzy sets take the place of manipulations of addition and multiplication for probabilities, respectively. Although the operations are somewhat formal, the obtained results seem interesting. A numerical example is attached to show a comparison of the proposed technique with the conventional one.     

Hybrid neural network shock absorber model

A hybrid neural network model is presented and described. The model is composed of a mechanical and thermodynamical part. The mechanical part is described by Akima spline in combination with a feed-forward neural network, while in the case of the thermodynamic part differential equation of dissipative heating is formulated. The interface between both part is provided by the neural network. To identify a proper parameter set of the hybrid model a shock absorber of a middle class passenger car is measured on a servo-hydraulic testing machine. As an excitation a stochastic signal with a predefined power spectral density (PSD) is used. Subsequently the hybrid shock absorber is implemented into a full vehicle model in ADAMS/Car to test its numerical performance and influence on the vertical vehicle dynamics. As reference a standard spline shock absorber model is taken. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

TRUST REGION METHOD IN NEURAL NETWORK

1.IntroductionHopfieldandTank[5]presentedamodeltosolvetravellingsalesmanproblem,thusinitiatingtheapplicationofneuralnetwork(NN)inthefieldofoptimization.SincethenmanyNNmodelshavebeenproposedtosolvelinearprogramming(LP)problems(13,8,11,14,15])andquadraticprogramming(oP)problems([1,8,20]),asLPandoPhavefundamentalimportanceinthetheoryandpracticeofoptimization.Therewerealsoafewmodelsforgeneralnonlinearprogramming(NP)problem([2,6,9,18]).InthispaperwewillpresentaHopfield-typeneuralnetworkmodelw…     

Bifurcation of a three-unit neural network

Considered is a system of delay differential equations modeling a time-delayed connecting network of three neurons without self-feedback. Discussing the change of the number of eigenvalues with zero real part, we locate the boundary of the stability region and finally determine the largest stability region of trivial solution. We investigate the existence of bifurcation phenomena of codimension one/two of the trivial equilibrium by considering the intersections of some parameter curves, which, in the aτ-half parameter plane, correspond to zero root or pure imaginary roots. In particular, the equivariant bifurcation is studied because of the equivariance of the system. We also present numerical simulations to demonstrate the rich dynamical behavior near the equivariant Pitchfork-Hopf bifurcation points, Hopf-Hopf bifurcation points, and some higher codimension bifurcation points.     

Univariate hyperbolic tangent neural network approximation

Here we study the univariate quantitative approximation of real and complex valued continuous functions on a compact interval or all the real line by quasi-interpolation hyperbolic tangent neural network operators. This approximation is derived by establishing Jackson type inequalities involving the modulus of continuity of the engaged function or its high order derivative. Our operators are defined by using a density function induced by the hyperbolic tangent function. The approximations are pointwise and with respect to the uniform norm. The related feed-forward neural network is with one hidden layer.     

Adaptive neural network model for time-series forecasting

In this study, a novel adaptive neural network (ADNN) with the adaptive metrics of inputs and a new mechanism for admixture of outputs is proposed for time-series prediction. The adaptive metrics of inputs can solve the problems of amplitude changing and trend determination, and avoid the over-fitting of networks. The new mechanism for admixture of outputs can adjust forecasting results by the relative error and make them more accurate. The proposed ADNN method can predict periodical time-series with a complicated structure. The experimental results show that the proposed model outperforms the auto-regression (AR), artificial neural network (ANN), and adaptive k-nearest neighbors (AKN) models. The ADNN model is proved to benefit from the merits of the ANN and the AKN through its’ novel structure with high robustness particularly for both chaotic and real time-series predictions.     

An unfeasibility view of neural network learning

We define the notion of a continuously differentiable perfect learning algorithm for multilayer neural network architectures and show that such algorithms do not exist provided that the length of the data set exceeds the number of involved parameters and the activation functions are logistic, tanh or sin.