• Journal of the Korean Data and Information Science Society
• /
• 제24권5호
• /
• pp.999-1011
• /
• 2013

본 연구에서는 논문이나 특허 등의 문서들의 인용 정보를 활용하여 연관성이 높고 중요한 특허를 추천하는 방법을 제안한다. 문서 간의 연관성 지표인 공통피인용횟수와 중요도 지표인 HITS를 적절한 형태로 결합한 뉴먼 커널로부터 두 정보의 반영 정도를 조율하는 것이 핵심이다. 제안하는 방법은 미래의 인용에 대한 예측 오차를 최소화하는 것으로 이를 통해 뉴먼 커널의 조율모수 ${\gamma}$를 적절하게 선택할 수 있다. 또한, 거대 인용 자료를 분석하기 위해 필요한 계산 기술에 대해서 자세히 논의한다. 마지막으로, 미국 등록 특허 400만 건에 대한 실증적 자료 분석을 시행한다.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제15권4호
• /
• pp.294-298
• /
• 2009

본 연구에서는 에지 성분의 추출을 보다 섬세하게 처리함으로써 이미지 화질을 개선하기 위한 방법을 제안한다. 제안 방법은 고전적인 보간을 응용한 ESII(Edge Sensitive Image Interpolation)으로 보간 커널 식에서 고정된 매개 변수를 사용하지 않고 주변 화소 값으로부터 적절한 정보를 얻어내서 매 화소마다 매개변수를 변화시킨다. LSE(Least Square Error)를 이용하여 CME(Camera Modelling Error)를 최소화 하도록 보간 할 화소 값을 결정함으로써 이미지를 복원하며, 기존 방법에 비교하여 객관적, 주관적 화질이 우수함을 실험, 제시하였으며 또한 영상을 1차원 정보로 분리하여 고려한 결과 계산상의 복잡도를 줄이는 효과를 기대할 수 있다.

• Structural Engineering and Mechanics
• /
• 제32권5호
• /
• pp.621-634
• /
• 2009

A numerical method is developed to investigate the effects of some geometric parameters and density variation on frequency characteristics of the circular and annular membranes with varying density. The discrete singular convolution method based on regularized Shannon's delta kernel is applied to obtain the frequency parameter. The obtained results have been compared with the analytical and numerical results of other researchers, which showed well agreement.

• 한국광학회지
• /
• 제23권6호
• /
• pp.246-250
• /
• 2012

We derive analytical expressions for the output spectral density and the noise power $P_{\beta}$ in noise loading analysis using Volterra kernels to characterize fiber nonlinearities. The bandwidth of the input noise source has little effect on $P_{\beta}$, but the power of the input noise source and the dispersion parameter value of the fiber have a significant effect on $P_{\beta}$. The Volterra method predicts ${\Delta}P_{\beta}[dB]$ = 30 dB/decade, which agrees very accurately over a wide range of fiber parameters compared with the numerical results by the split-step Fourier method. Therefore the Volterra method could be useful to predict the performance of a dense WDM system when we plan to upgrade fiber or increase signal power.

• 에너지공학
• /
• 제23권3호
• /
• pp.1-6
• /
• 2014

The more or less homogeneous fuel-air mixture that exists at the end of the compression process is ignited by an electric ignition spark from a spark plug shortly before top dead center. The actual moment of ignition is an optimization parameter; it is adapted to the engine operation so that an optimum combustion process is obtained. Brake mean effective pressure (BMEP) of the spark ignition energy control device (IECD) than conventional spark system at the stoichiometric mixture is increased about 9%. For lean burn engine, the lean limit is extended about 25% by using the IECD. It was considered the stability of combustion by the increase of flame kernel according to the high ignition energy supplies in initial period and discharge energy period lengthen by using the IECD.

• Structural Engineering and Mechanics
• /
• 제64권1호
• /
• pp.71-79
• /
• 2017

In this study, interaction of several interface cracks located between a functionally graded material (FGM) layer and an elastic layer under anti-plane deformation based on the distributed dislocation technique (DDT) is analyzed. The variation of the shear modulus of the functionally graded coating is modeled by an exponential and linear function along the thickness of the layer. The complex Fourier transform is applied to governing equation to derive a system of singular integral equations with Cauchy type kernel. These equations are solved by a numerical method to obtain the stress intensity factors (SIFs) at the crack tips. The effects of non-homogeneity parameters for exponentially and linearly form of shear modulus, the thickness of the layers and the length of crack on the SIFs for several interface cracks are investigated. The results reveal that the magnitude of SIFs decrease with increasing of FG parameter and thickness of FGM layer. The values of SIFs for FGM layer with exponential form is less than the linear form.

• Journal of the Korean Data and Information Science Society
• /
• 제20권5호
• /
• pp.925-931
• /
• 2009

In this paper we propose a variance function estimation method for replicated data based on averages of squared residuals obtained from estimated mean function by the least squares support vector machine. Newton-Raphson method is used to obtain associated parameter vector for the variance function estimation. Furthermore, the cross validation functions are introduced to select the hyper-parameters which affect the performance of the proposed estimation method. Experimental results are then presented which illustrate the performance of the proposed procedure.

• ETRI Journal
• /
• 제38권3호
• /
• pp.540-550
• /
• 2016

Spectral clustering is a powerful tool for exploratory data analysis. Many existing spectral clustering algorithms typically measure the similarity by using a Gaussian kernel function or an undirected k-nearest neighbor (kNN) graph, which cannot reveal the real clusters when the data are not well separated. In this paper, to improve the spectral clustering, we consider a robust similarity measure based on the shared nearest neighbors in a directed kNN graph. We propose two novel algorithms for spectral clustering: one based on the number of shared nearest neighbors, and one based on their closeness. The proposed algorithms are able to explore the underlying similarity relationships between data points, and are robust to datasets that are not well separated. Moreover, the proposed algorithms have only one parameter, k. We evaluated the proposed algorithms using synthetic and real-world datasets. The experimental results demonstrate that the proposed algorithms not only achieve a good level of performance, they also outperform the traditional spectral clustering algorithms.

• Journal of Mechanical Science and Technology
• /
• 제16권3호
• /
• pp.395-402
• /
• 2002

This paper shows how to formulate the transient analysis of 2-dimensional Hagen-Poiseuille flow using smoothed particle hydrodynamics (SPH). Treatments of viscosity, particle approximation and boundary conditions are explained. Numerical tests are calculated to examine effects caused by the number of particles, the number of particles per smoothing length, artificial viscosity and time increments for 2-dimensional Hagen-Poiseuille flow. Artificial viscosity for reducing the numerical instability directly affects the velocity of the flow, though effects of the other parameters do not produce as much effect as artificial viscosity. Numerical solutions using SPH show close agreement with the exact ones for the model flow, but SPH parameter must be chosen carefully Numerical solutions indicate that SPH is also an effective method for the analysis of 2-dimensional Hagen-Poiseuille flow.

• 제어로봇시스템학회논문지
• /
• 제9권9호
• /
• pp.658-663
• /
• 2003

In deterministic design of feedback controllers for uncertain dynamic systems, the upper bound of the uncertainty is very important to guarantee the stability of the closed loop system. In this paper, we assume that the upper bound of the uncertainty is formulated using a Fredholm integral equation of the first kind, that is, an integral of the product of a predefined kernel with an unknown influence function. We propose an adaptation law that is capable of estimating this upper bound. Using this adaptive upper bound, we design an adaptive variable structure control (AVSC), which guarantees asymptotic stability/ultimate boundedness of uncertain dynamic systems. The illustrative example shows the proposed AVSC is effective for uncertain dynamic systems.