• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2946-2948
• /
• 2000

In this paper, we proposed the Fuzzy Polynomial Neural Networks(FPNN) model with fuzzy activation node. The proposed FPNN structure is generated from the mutual combination of PNN(Polynomial Neural Networks) structure and fuzzy inference system. The premise of fuzzy inference rules defines by triangular and gaussian type membership function. The fuzzy inference method uses simplified and regression polynomial inference method which is based on the consequence of fuzzy rule expressed with a polynomial such as linear, quadratic and modified quadratic equation are used. The structure of FPNN is not fixed like in conventional Neural Networks and can be generated. The design procedure to obtain an optimal model structure utilizing FPNN algorithm is shown in each stage. Gas furnace time series data used to evaluate the performance of our proposed model.

• Proceedings of the KIEE Conference
• /
• 1998.11b
• /
• pp.675-677
• /
• 1998

In this paper, we propose the fuzzy inference algorithm with multi-layer structure. MFIS(Multi-layer Fuzzy Inference System) uses PNN(Polynomial Neural networks) structure and the fuzzy inference method. The PNN is the extended structure of the GMDH(Group Method of Data Hendling), and uses several types of polynomials such as linear, quadratic and cubic, as well as the biquadratic polynomial used in the GMDH. In the fuzzy inference method, the simplified and regression polynomial inference methods are used. Here, the regression polynomial inference is based on consequence of fuzzy rules with the polynomial equations such as linear, quadratic and cubic equation. Each node of the MFIS is defined as fuzzy rules and its structure is a kind of neuro-fuzzy structure. We use the training and testing data set to obtain a balance between the approximation and the generalization of process model. Several numerical examples are used to evaluate the performance of the our proposed model.

• Journal of the Korean Society of Safety
• /
• v.30 no.1
• /
• pp.8-13
• /
• 2015

The experimental design methodology was applied in the drop tube furnace (DTF) to predict the various combustion properties according to the operating conditions and to assess the coal plant safety. Response surface method (RSM) was introduced as a design of experiment, and the database for RSM was set with the numerical simulation of DTF. The dependent variables such as burnout ratios (BOR) of coal and $CO/CO_2$ ratios were mathematically described as a function of three independent variables (coal particle size, carrier gas flow rate, wall temperature) being modeled by the use of the central composite design (CCD), and evaluated using a second-order polynomial multiple regression model. The prediction of BOR showed a high coefficient of determination (R2) value, thus ensuring a satisfactory adjustment of the second-order polynomial multiple regression model with the simulation data. However, $CO/CO_2$ ratio had a big difference between calculated values and predicted values using conventional RSM, which might be mainly due to the dependent variable increses or decrease very steeply, and hence the second order polynomial cannot follow the rates. To relax the increasing rate of dependent variable, $CO/CO_2$ ratio was taken as common logarithms and worked again with RSM. The application of logarithms in the transformation of dependent variables showed that the accuracy was highly enhanced and predicted the simulation data well.

• ETRI Journal
• /
• v.26 no.4
• /
• pp.297-306
• /
• 2004

A plasma is a collection of charged particles and on average is electrically neutral. In fabricating integrated circuits, plasma etching is a key means to transfer a photoresist pattern into an underlayer material. To construct a predictive model of plasma-etching processes, a polynomial neural network (PNN) is applied. This process was characterized by a full factorial experiment, and two attributes modeled are its etch rate and DC bias. According to the number of input variables and type of polynomials to each node, the prediction performance of the PNN was optimized. The various performances of the PNN in diverse environments were compared to three types of statistical regression models and the adaptive network fuzzy inference system (ANFIS). As the demonstrated high-prediction ability in the simulation results shows, the PNN is efficient and much more accurate from the point of view of approximation and prediction abilities.

• Communications for Statistical Applications and Methods
• /
• v.5 no.1
• /
• pp.55-67
• /
• 1998

Recent work by Shin and Sarkar (1996) examines model misspecification problems for nonstationary time series. Shin and Sarkar introduce a general regression model with integrated errors and one system of integrated regressors and discuss the limiting distributions of the OLS estimators and the usual OLS statistics such as $\hat{\sigma^2}$t, DW and $R^2$. We analyze three different model misspecification problems through a Monte Carlo study and investigate each model misspecification problem. Our Monte Carlo experiments show that DW and $R^2$ can be in general used as diagnostic tools to detect spurious regression, misspecification of nonstationary autoregressive and polynomial regression models.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.1
• /
• pp.131-141
• /
• 2003

In this paper, we prove that multi-variate fuzzy polynomials are universal approximators for multi-variate fuzzy functions which are the extension principle of continuous real-valued function under $T_W-based$ fuzzy arithmetic operations for a distance measure that Buckley et al.(1999) used. We also consider a class of fuzzy polynomial regression model. A mixed non-linear programming approach is used to derive the satisfying solution.

• ETRI Journal
• /
• v.27 no.6
• /
• pp.747-758
• /
• 2005

The problem of selection among competing models has been a fundamental issue in statistical data analysis. Good fits to data can be misleading since they can result from properties of the model that have nothing to do with it being a close approximation to the source distribution of interest (for example, overfitting). In this study we focus on the preference among models from a family of polynomial regressors. Three decades of research has spawned a number of plausible techniques for the selection of models, namely, Akaike's Finite Prediction Error (FPE) and Information Criterion (AIC), Schwartz's criterion (SCH), Generalized Cross Validation (GCV), Wallace's Minimum Message Length (MML), Minimum Description Length (MDL), and Vapnik's Structural Risk Minimization (SRM). The fundamental similarity between all these principles is their attempt to define an appropriate balance between the complexity of models and their ability to explain the data. This paper presents an empirical study of the above principles in the context of model selection, where the models under consideration are univariate polynomials. The paper includes a detailed empirical evaluation of the model selection methods on six target functions, with varying sample sizes and added Gaussian noise. The results from the study appear to provide strong evidence in support of the MML- and SRM- based methods over the other standard approaches (FPE, AIC, SCH and GCV).

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.337-339
• /
• 2004

In this paper, we discuss optimal design of Fuzzy Polynomial Neural Networks by means of Genetic Algorithms(GAs). Proceeding the layer, this model creates the optimal network architecture through the selection and the elimination of nodes by itself. So, there is characteristic of flexibility. We use a triangle and a Gaussian-like membership function in premise part of rules and design the consequent structure by constant and regression polynomial (linear, quadratic and modified quadratic) function between input and output variables. GAs is applied to improve the performance with optimal input variables and number of input variables and order. To evaluate the performance of the GAs-based FPNNs, the models are experimented with the use of Medical Imaging System(MIS) data.

• Proceedings of the KIEE Conference
• /
• 2006.04a
• /
• pp.270-272
• /
• 2006

In this paper, we discuss optimal design of Fuzzy Polynomial Neural Networks by means of Genetic Algorithms(GAs) using symbolic coding for non-linear data. One of the major subject of genetic algorithms is representation of chromosomes. The proposed model optimized by the means genetic algorithms which used symbolic code to represent chromosomes. The proposed gFPNN used a triangle and a Gaussian-like membership function in premise part of rules and design the consequent structure by constant and regression polynomial (linear, quadratic and modified quadratic) function between input and output variables. The performance of the proposed model is quantified through experimentation that exploits standard data already used in fuzzy modeling. These results reveal superiority of the proposed networks over the existing fuzzy and neural models.

• Korean Journal of Health Education and Promotion
• /
• v.12 no.2
• /
• pp.36-47
• /
• 1995

This research was obtained from analyzing how the physiques of the 3rd grade students of high school for males and females and developed for the last eleven years(from 1983 to 1993). By the physiques and nutritional index of physical growth and development, Relative Body Weight of 36.62 exceeded the standard, on the other hand females showed lower records than the standard. Relative Chest Girth Index belonged to the normal type of males and females in all, in the comparison of the records between 1983 and 1993, males increased in average 0.29 and females in average 0.55. Relative Chest Girth Index of females was greater than that of females. By the results of Relative Sitting Height Index, growth of the lower body for males and females was greater than that of males. In case of Vervaeck Index, males increased in average 2.04 but females increased in average 1, 20 relatively less than males. These phenomena provided for the evidence of the deficient nutrition in females. In the regression models of body height and body weight within a certain period, statistical regression model types which best indicated chronological average changes of body height and body weight, took 3rd Order Polynomial Regression Model rather than linear regression model. In females, statistical regression model types which best is suitable for chronological average change of body height and body weight, took 4th and 2nd Order Polynomial Regression Model respectively. The prediction value of 1995 by estimated polynomial regression model anticipated that body height of 3rd grade year students of high school of males in 1993 went on increasing from 170.87cm to 171.79cm in average 0.92cm growth and that of females from 158.99cm to 160.79cm in average 1.80cm growth. In addition, body weight of males seemed to increase from 62.58kg to 64.52kg in average 1.94kg growth and that of females seemed to increase from 54.05kg to 54.19kg in average 0.14kg growth. Linear Regression Model was suitable for the regression model of body weight for body height. Prediction on increase of an average body weight for body height was that, according to growth of body height 1cm in males, body weight increased 1.41kg averagely and that of females 0.86kg. For that reason, we came to conclusion that body weight increase for body height 1cm in males was greater than that in females on average.