• Investigative Magnetic Resonance Imaging
• /
• v.2 no.1
• /
• pp.58-66
• /
• 1998

Purpose : This paper introduces a new three dimensional magnetic Resonance Image classification which is based on Mar kov Random Field-Gibbs Random Field with a line model. Material and Methods : The performance of the Gibbs Classifier over a statistically heterogeneous image can be improved if the local stationary regions in the image are disassociated from each other through the mechanism of the interaction parameters defined at the local neighborhood level. This usually involves the construction of a line model for the image. In this paper we construct a line model for multisignature images based on the differential of the image which can provide an a priori estimate of the unobservable line field, which may lie in regions with significantly different statistics. the line model estimated from the original image data can in turn be used to alter the values of the interaction parameters of the Gibbs Classifier. Results : MRF-Gibbs classifier for volumetric MR images is developed under the condition that the domain of the image classification is $E^{3}$ space rather thatn the conventional $E^{2}$ space. Compared to context free classification, MRF-Gibbs classifier performed better in homogeneous and along boundaries since contextual information is used during the classification. Conclusion : We construct a line model for multisignature, multidimensional image and derive the interaction parameter for determining the energy function of MRF-Gibbs classifier.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.23 no.7 s.166
• /
• pp.1139-1146
• /
• 1999

Reliability analysis of structures based on fracture mechanics requires knowledge on statistical characteristics of the parameter C and m in the fatigue crack growth law, $da/dN=C({\Delta}K)^m$. The purpose of the present study is to investigate if it is possible to predict fatigue crack growth rate by only the fluctuation of the parameter C. In this study, Paris-Erdogan law is adopted, where the author treat the parameter C as random and m as constant. The fluctuation of crack growth rate is assumed only due to the parameter C. The growth resistance coefficient of material to fatigue crack growth (Z=1/C) was treated as a spatial stochastic process, which varies randomly on the crack path. The theoretical crack growth rates at various stress intensity factor range are discussed. Constant ${\Delta}K$ fatigue crack growth tests were performed on the structural steel, SM45C. The experimental data were analyzed to determine the autocorrelation function and Weibull distributions of the fatigue crack growth resistance. And also, the effect of the parameter m of Paris' law due to variation of fatigue crack growth resistance was discussed.

• The Korean Journal of Applied Statistics
• /
• v.29 no.1
• /
• pp.257-266
• /
• 2016

Random Coefficient Autoregressive models (RCA) have attracted increased interest due to the wide range of applications in biology, economics, meteorology and finance. We consider an RCA as an appropriate model for non-linear properties and better than an AR model for linear properties. We study the methods of RCA parameter estimation. Especially we proposed the special case that an random coefficient ${\phi}(t)$ has the initial value ${\phi}(0)$ in the RCA model. In practical study, we estimated the parameters and compared Prediction Error Sum of Squares (PRESS) criterion between AR and RCA using Korean Mumps data.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.54 no.3
• /
• pp.55-63
• /
• 2012

Soil properties are not random values which is represented by mean and standard deviation but show spatial correlation. Especially, soils are highly variable in their properties and rarely homogeneous. Thus, the accuracy and reliability of probabilistic analysis results is decreased when using only one random variable as design parameter. In this paper, to consider spatial variability of soil property, one-dimensional random fields of coefficient of consolidation ($C_v$) were generated based on a Karhunen-Loeve expansion. A Latin hypercube Monte Calro simulation coupled with finite difference method for Terzaghi's one dimensional consolidation theory was then used to probabilistic analysis. The results show that the failure probability is smaller when consider spatial variability of $C_v$ than not considered and the failure probability increased when the autocorrelation distance increased. Thus, the uncertainty of soil can be overestimated when spatial variability of soil property is not considered, and therefore, to perform a more accurate probabilistic analysis, spatial variability of soil property needed to be considered.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.3
• /
• pp.285-291
• /
• 2022

The number of people enrolling in universities is rising due to the simplicity of applying and the benefit of earning a bachelor's degree. However, the on-time graduation rate has declined since plenty of students fail to complete their courses and take longer to get their diplomas. Even though there are various reasons leading to the aforementioned problem, it is crucial to emphasize the cause originating from the management and care of learners. In fact, understanding students' difficult situations and offering timely Number of Test data and advice would help prevent college dropouts or graduate delays. In this study, we present a machine learning-based method for early detection at-risk students, using data obtained from graduates of the Faculty of Information Technology, Dainam University, Vietnam. We experiment with several fundamental machine learning methods before implementing the parameter optimization techniques. In comparison to the other strategies, Random Forest and Grid Search (RF&GS) and Random Forest and Random Search (RF&RS) provided more accurate predictions for identifying at-risk students.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.28 no.2
• /
• pp.146-151
• /
• 2005

Some distributions have been used for diagnosing the lead time demand distribution in inventory system. In this paper, we describe the negative binomial distribution as a suitable demand distribution for a specific retail inventory management application. We here assume that customer order sizes are described by the Poisson distribution with the random parameter following a gamma distribution. This implies in turn that the negative binomial distribution is obtained by mixing the mean of the Poisson distribution with a gamma distribution. The purpose of this paper is to give an interpretation of the negative binomial demand process by considering the sources of variability in the unknown Poisson parameter. Such variability comes from the unknown demand rate and the unknown lead time interval.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.58 no.2
• /
• pp.164-171
• /
• 2009

This paper presents new learning algorithm of dynamic Bayesian networks (DBN) by means of constrained least square (LS) estimation algorithm and gradient descent method. First, we propose constrained LS based parameter estimation for a Markov chain (MC) model given observation data sets. Next, a gradient descent optimization is utilized for online estimation of a hidden Markov model (HMM), which is bi-linearly constructed by adding an observation variable to a MC model. We achieve numerical simulations to prove its reliability and superiority in which a series of non stationary random signal is applied for the DBN models respectively.

• Journal of information and communication convergence engineering
• /
• v.9 no.5
• /
• pp.539-544
• /
• 2011

In this paper, a parameter optimization method of RIO-DC (RED (Random Early Detection) with In and Out-De-Coupled Queues) scheme for Assured Service (AS) in Differentiated Services (DiffServ) is proposed. In order to optimize QoS (Quality of Service) performance of the RIO-DC policy for AS in terms of maximum tolerable latency, link utilization, fairness, etc., we should design router nodes with proper RIO-DC operating parameter values. Therefore, we propose a RIO-DC configuration method and the admission control criterion, considering the allocated bandwidth to each subclass and the corresponding buffer size, to increase throughput for In-profile traffic and link utilization. Simulation results show that RIO-DC with the proposed parameter values guarantees QoS performance comparable with the RIO scheme and it improves fairness between AS flows remarkably.

• Korea-Australia Rheology Journal
• /
• v.12 no.3_4
• /
• pp.175-179
• /
• 2000

Rheological properties of chitosan solutions were investigated as a function of polymer concentration. The viscosity curves for chitosan solutions consisted of two distinct viscosity regions, the Newtonian zero-shear viscosity (η$_{0}$) region and the shear rate dependent apparent viscosity (η$_{app}$) region. The shear rate dependence of viscosity was more clearly observed at higher chitosan concentrations. The critical coil overlap parameter (C＊〔η〕) was determined to be approximately 3.2 from a plot of zero-shear specific viscosity η$_{sp,0}$ vs coil overlap parameter (C〔η〕), which was lower than C〔η〕4.0 reported for other random coil polysaccharides. It was also found that the slope of η$_{sp,0}$ vs C〔η〕 was 3.9 at concentrated C〔η〕>C＊〔η〕domain, while 1.2 at dilute C〔η〕$_{0}$ ${\gamma}$/${\gamma}$$_{0.8}$ relation.ion.n.n.

• Journal of the Korean Statistical Society
• /
• v.23 no.1
• /
• pp.13-32
• /
• 1994

Let A(n) and B(n) be sequences of $m \times m$ random matrices with a joint asymptotic distribution as $n \to \infty$. The asymptotic distribution of the ordered roots of $$\mid$A(n) - f B(n)$\mid$ = 0$ depends on the multiplicity of the roots of a determinatal equation involving parameter roots. This paper treats the asymptotic distribution of the roots of the above determinantal equation in the case where some of parameter roots are zero. Furthermore, we apply our results to deriving the asymptotic distributions of the eigenvalues of the MANOVA matrix in the noncentral case when the underlying distribution is not multivariate normal and some parameter roots are zero.