• 한국지능시스템학회논문지
• /
• 제17권2호
• /
• pp.196-201
• /
• 2007

Fuzzy C-Means (FCM) 알고리즘은 probabilitic 멤버쉽을 사용하는 클러스터링 방법으로서 널리 쓰이고 있다. 하지만 이 방법은 노이즈에 대하여 민감한 성질을 가진다는 단점이 있다. 따라서 본 논문에서는 이러한 노이즈에 민감한 성질을 보완하기 위해서 데이터의 밀도추정을 이용하여 새로운 FCM 알고리즘을 제안한다. 본 논문에서 제안된 알고리즘은 FCM과 비슷한 성능의 클러스터링 수행이 가능하며, 노이즈가 포함된 데이터에서는 FCM보다 더 나은 성능을 보여준다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.408-413
• /
• 2005

A new methodology for discrete non-Gaussian probability density estimation is investigated in this paper based on a dynamic Bayesian network (DBN) and kernel functions. The estimator consists of a DBN in which the transition distribution is represented with kernel functions. The estimator parameters are determined through a recursive learning algorithm according to the maximum likelihood (ML) scheme. A discrete-type Poisson distribution is generated in a simulation experiment to evaluate the proposed method. In addition, an unknown probability density generated by nonlinear transformation of a Poisson random variable is simulated. Computer simulations numerically demonstrate that the method successfully estimates the unknown probability distribution function (PDF).

• Communications for Statistical Applications and Methods
• /
• 제4권3호
• /
• pp.833-845
• /
• 1997

Fourier expansion is considered for the deconvolution problem of estimating a probability density function when the sample observations are contaminated with random noise. In the log-Fourier method of density estimation for data without noise, the logarithm of the unknown density function is approximated by a trigonometric function, the unknown parameters of which are estimated by maximum likelihood. The log-Fourier density estimation method, which has been considered theoretically by Koo and Chung (1997), is studied for the finite-sample case with noise. Numerical examples using simulated data are given to show the performance of the log-Fourier deconvolution.

• 정보처리학회논문지A
• /
• 제9A권1호
• /
• pp.75-82
• /
• 2002

전력 소모 예측에 필요한 신호 전이 밀도를 구하기 위하여, 제로 지연 모델에 대한 견실한 신호 전이 밀도 전파 방법이 제시된다. 제로 기연 모델을 위한 전력 예측은 전력 소모의 하한 경계값을 위한 적절한 기준이다. 입력 특성이 일반적으로 설계 단계에 알려져 있지 않기 때문에 광범위한 입력 특성에 대한 견실한 예측은 전력 소모에 대하여 매우 중요하다. 본 연구에서는 기존의 신호 전이 예측 방법에 대하여 입력 및 출력의 변이 특성을 분석하고 이러한 분석 결과에 근거하여 새로운 견실한 신호 전이 밀도 전파 방법을 제안한다. 실제 회로에 적용하기 위하여 전력 예측의 정확성에 크게 영향을 미치는 재수렴성 경로를 고려한 알고리즘을 제안 및 연구한다. 실험에 의하면 제안한 방법이 기존의 방법과 비교할 때 더욱 양호한 견실성 및 종래의 방식에 상응하는 정확성과 경과 시간을 보여준다.

• 한국지반공학회:학술대회논문집
• /
• 한국지반공학회 2010년도 춘계 학술발표회
• /
• pp.717-721
• /
• 2010

Soil volume conversion factors are used for estimation of an excavated the soil volume which will be removed or added in levelling the ground surface of a construction site. An accurate evaluation method will help us reduce a construction cost and time consuming. In this study, we performed the laboratory tests, including grain size measurement, water content, specific gravity, porosity, density and XRD tests, to suggest reliable soil volume conversion factors and weathering indices in field using nondestructive methods. The weathering index and soil volume conversion factor L are obtained for different types of soils. At results, the CIW index is the best method measuring the weathering index and the factor L is relative to natural porosity, void ratio, density and dry density.

• Nuclear Engineering and Technology
• /
• 제48권5호
• /
• pp.1280-1290
• /
• 2016

This paper presents an advanced estimation method for obtaining the probability density functions of a damage parameter for valve leakage detection in a reciprocating pump. The estimation method is based on a comparison of model data which are simulated by using a mathematical model, and experimental data which are measured on the inside and outside of the reciprocating pump in operation. The mathematical model, which is simplified and extended on the basis of previous models, describes not only the normal state of the pump, but also its abnormal state caused by valve leakage. The pressure in the cylinder is expressed as a function of the crankshaft angle, and an additional volume flow rate due to the valve leakage is quantified by a damage parameter in the mathematical model. The change in the cylinder pressure profiles due to the suction valve leakage is noticeable in the compression and expansion modes of the pump. The damage parameter value over 300 cycles is calculated in two ways, considering advance or delay in the opening and closing angles of the discharge valves. The probability density functions of the damage parameter are compared for diagnosis and prognosis on the basis of the probabilistic features of valve leakage.

• 한국컴퓨터정보학회논문지
• /
• 제17권7호
• /
• pp.45-52
• /
• 2012

본 논문에서는 다중 클래스 아다부스트 기반의 분류기를 이용하여 엘리베이터 내 군집 밀도를 추정하는 방법을 제안한다. SOM을 사용하는 기존의 방법은 재현성이 떨어지며 충분한 성능을 내지 못한다. 제안한 방법은 GLDM(Grey-Level Dependency Matrix)과 GGDM(Grey-Gradient Dependency Matrix)의 텍스처 특징과 다중 클래스 아다부스트 기반의 분류기를 통해 실내 군집 밀도를 추정한다. 다중 클래스를 분류하기 위해 기존의 아다부스트 알고리즘에서 웨이트 업데이트 식을 변형하여 더 높은 성능의 약한 분류기를 생성하도록 하였다. 군집 밀도는 인원수에 따라 0명, 1~2명, 3~4명, 5명 이상 등 네 가지 클래스로 구분하였다. 엘리베이터 내 영상을 이용한 모의 실험 결과 제안된 방법은 기존의 방법보다 약 20% 정도의 검출률 향상을 나타내었다.

• Communications for Statistical Applications and Methods
• /
• 제19권4호
• /
• pp.517-526
• /
• 2012

In this paper a support vector method is proposed for use when the sample observations are contaminated by a normally distributed measurement error. The performance of deconvolution density estimators based on the support vector method is explored and compared with kernel density estimators by means of a simulation study. An interesting result was that for the estimation of kurtotic density, the support vector deconvolution estimator with a Gaussian kernel showed a better performance than the classical deconvolution kernel estimator.

• Journal of Forest and Environmental Science
• /
• 제33권4호
• /
• pp.305-314
• /
• 2017

This study aimed to predict changes in forest area using a probability density function, in order to promote effective forest management in the area north of the civilian control line (known as the Minbuk area) in Korea. Time series analysis (2010 and 2016) of forest area using land cover maps and accessibility expressed by distance covariates (distance from buildings, roads, and civilian control line) was applied to a probability density function. In order to estimate the probability density function, mean and variance were calculated using three methods: area weight (AW), area rate weight (ARW), and sample area change rate weight (SRW). Forest area increases in regions with lower accessibility (i.e., greater distance) from buildings and roads, but no relationship with accessibility from the civilian control line was found. Estimation of forest area change using different distance covariates shows that SRW using distance from buildings provides the most accurate estimation, with around 0.98-fold difference from actual forest area change, and performs well in a Chi-Square test. Furthermore, estimation of forest area until 2028 using SRW and distance from buildings most closely replicates patterns of actual forest area changes, suggesting that estimation of future change could be possible using this method. The method allows investigation of the current status of land cover in the Minbuk area, as well as predictions of future changes in forest area that could be utilized in forest management planning and policymaking in the northern area.

• Communications for Statistical Applications and Methods
• /
• 제13권3호
• /
• 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.