• Proceedings of the KIEE Conference
• /
• 1998.07b
• /
• pp.519-521
• /
• 1998

Fuzzy inference systems have found many applications in recent years. The fuzzy inference system design procedure is related to an expert or a skilled human operator in many fields. Various attempts have been made in optimizing its structure using genetic algorithm automated designs. This paper presents a new approach to structurally optimized designs of FNN models. The messy genetic algorithm is used to obtain structurally optimized fuzzy neural network models. Structural optimization is regarded important before neural network based learning is switched into. We have applied the method to the problem of a time series estimation.

• Journal of the Korean earth science society
• /
• v.35 no.5
• /
• pp.362-374
• /
• 2014

This study investigated the initial mass function (IMF) and star formation history of high-mass stars in the Small Magellanic Cloud (SMC) using a population synthesis technique. We used the photometric survey catalog of Lee (2013) as the observable quantities and compare them with those of synthetic populations based on Bayesian inference. For the IMF slope (${\Gamma}$) range of -1.1 to -3.5 with steps of 0.1, five types of star formation models were tested: 1) continuous; 2) single burst at 10 Myr; 3) single burst at 60 Myr; 4) double bursts at those epochs; and 5) a complex hybrid model. In this study, a total of 125 models were tested. Based on the model calculations, it was found that the continuous model could simulate the high-mass stars of the SMC and that its IMF slope was -1.6 which is slightly steeper than Salpeter's IMF, i.e., ${\Gamma}=-1.35$.

• Journal of the Korean Data and Information Science Society
• /
• v.10 no.1
• /
• pp.119-133
• /
• 1999

A Markov Chain Monte Carlo method is developed to compute the software reliability model. We consider computation problem for determining of posterior distibution in Bayseian inference. Metropolis algorithms along with Gibbs sampling are proposed to preform the Bayesian inference of the Mixed model with record value statistics. For model determiniation, we explored the prequential conditional predictive ordinate criterion that selects the best model with the largest posterior likelihood among models using all possible subsets of the component intensity functions. To relax the monotonic intensity function assumptions. A numerical example with simulated data set is given.

• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2314-2316
• /
• 2000

In the construction of successful fuzzy models and/or controllers for nonlinear systems, identification of a good fuzzy Neural inference system is an important yet difficult problem, which is traditionally accomplished by trial and error process. In this paper, we propose a systematic identification procedure for complex multi-input single- output nonlinear systems with DNA coding method.DNA coding method is optimization algorithm based on biological DNA as are conventional genetic algothms (GAs). We also propose a new coding method for applying the DNA coding method to the identification of fuzzy Neural models. To acquire optimal TS fuzzy model with higher accuracy and economical size, we use the DNA coding method to optimize the parameters and the number of fuzzy inference system.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.3
• /
• pp.852-861
• /
• 2000

Software reliability growth models are used in testing stages of software development to model the error content and time intervals betwewn software failures. In this paper, using priors for the number of fault with the negative binomial distribution nd the error rate with gamma distribution, Bayesian inference and model selection method for Jelinski-Moranda and Goel-Okumoto and Schick-Wolverton models in software reliability. For model selection, we explored the sum of the relative error, Braun statistic and median variation. In Bayesian computation process, we could avoid the multiple integration by the use of Gibbs sampling, which is a kind of Markov Chain Monte Carolo method to compute the posterior distribution. Using simulated data, Bayesian inference and model selection is studied.

• Communications for Statistical Applications and Methods
• /
• v.9 no.1
• /
• pp.115-128
• /
• 2002

Hierarchical models are widely used for inference on correlated parameters as a compromise between underfitting and overfilling problems. In this paper, we take a Bayesian approach to analyzing hierarchical models and suggest a Markov chain Monte Carlo methods to get around computational difficulties in Bayesian analysis of the hierarchical models. We apply the method to a real data on smoking and lung cancer which are collected from cities in China.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.120-123
• /
• 1997

This paper presents models of fuzzy inference systems that can be built from a set of input-output training data pairs through hybrid structure-parameter learning. Fuzzy inference systems has the difficulty of parameter learning. Here we develop a coding format to determine a fuzzy neural network(FNN) model by chromosome in a genetic algorithm(GA) and present systematic approach to identify the parameters and structure of FNN. The proposed FNN can automatically identify the fuzzy rules and tune the membership functions by modifying the connection weights of the networks using the GA and the back-propagation learning algorithm. In order to show effectiveness of it we simulate and compare with conventional methods.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.340-342
• /
• 2004

In this paper, we introduce a new category of fuzzy inference systems baled on information granulation to carry out the model identification of complex and nonlinear systems. Informal speaking, information granules are viewed as linked collections of objects(data, in particular) drawn together by the criteria of proximity, similarity, or functionality. Granulation of information with the aid of Hard C-Means(HCM) clustering algorithm help determine the initial parameters of fuzzy model such as the initial apexes of the membership functions and the initial values of polyminial functions being used in the premise and consequence part of the fuzzy rules. And the initial parameters are tuned effectively with the aid of the genetic algorithms(GAs) and the least square method. The proposed model is contrasted with the performance of the conventional fuzzy models in the literature.

• Journal of Institute of Control, Robotics and Systems
• /
• v.7 no.11
• /
• pp.958-961
• /
• 2001

Control of Industrial processes is very difficult due to nonlinear dynamics, effect of disturbances and modeling errors. M.Morari proposed Internal Model Control(IMC) system that can be effectively applied to the systems with model uncertainties and time delays. The advantage of IMC is their robustness with respect to a model mismatch and disturbances. But it is difficult to apply for nonlinear systems. ANFIS(Adaptive Neuro-Fuzzy Inference System) which contains multiple linear models as consequent part is used to model nonlinear systems. Generally, the linear parameters in ANFIS can be effectively utilized to control a nonlinear systems. In this paper, we propose new ANFIS-based IMC controller for nonlinear systems. Numerical simulation results show that the proposed control scheme has good performances.

• Communications for Statistical Applications and Methods
• /
• v.13 no.3
• /
• pp.503-512
• /
• 2006

In this paper, the Inverse Weibull distribution parameters have been estimated using a new estimation technique based on the non-parametric kernel density function that introduced as an alternative and reliable technique for estimation in life testing models. This technique will require bootstrapping from a set of sample observations for constructing the density functions of pivotal quantities and thus the confidence intervals for the distribution parameters. The performances of this technique have been studied comparing to the conditional inference on the basis of the mean lengths and the covering percentage of the confidence intervals, via Monte Carlo simulations. The simulation results indicated the robustness of the proposed method that yield reasonably accurate inferences even with fewer bootstrap replications and it is easy to be used than the conditional approach. Finally, a numerical example is given to illustrate the densities and the inferential methods developed in this paper.