• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05a
• /
• pp.309-312
• /
• 2003

추천 시스템을 위한 여과 기술에는 협력적 여과, 내용기반 여과 등이 있다. 협력적 여과 방법은 적용이 용이한 반면 회소성 문제와 초기 평가 문제가 있으며, 내용기반 여과는 정보의 질을 구분하는 것이 어려워 효과가 적다는 단점이 있다. 신경망 기반 협력적 여과 방법은 이러한 문제를 해결하고 있지만, 사용자의 수가 많아지면 모델이 커져 효율성이 떨어지는 문제가 있다. 본 논문에서는 신경망 기반 협력적 여과의 효율성을 높이기 위해 상관도를 고려하는 신경망 기반 협력적 여과를 제안한다. 여기서 상관도란 피어슨 상관계수를 이용하여 구해진 상관계수의 절대값을 의미하며 상관도가 높다라는 것은 상관계수의 절대값이 1에 가까운 경우를 말한다. 본 논문에서는 EachMovie 데이터를 이용하여 제안한 방법의 우수함을 보인다.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2003.08a
• /
• pp.198-198
• /
• 2003

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.1
• /
• pp.105-111
• /
• 2013

We have developed an analog associative memory implemented with an analog array which has linear writing and erasing characteristics. The associative memory adopts a winner-take-all strategy. The operation for reading in the memory is executed with an absolute differencing circuit and a winner-take-all (WTA) circuit suitable for a nearest-match function of a content-addressable memory. We also present a system architecture that enables highly-paralleled fast writing and quick readout as well as high integration density. A multiple memory cell configuration is also presented for achieving higher integration density, quick readout, and fast writing. The system technology presented here is ideal for a real time recognition system. We simulate the function of the mechanism by menas of Hspice with $1.2{\mu}$ double poly CMOS parameters of MOSIS fabrication process.

• The Korean Journal of Applied Statistics
• /
• v.28 no.6
• /
• pp.1245-1255
• /
• 2015

In this paper, we investigated the major 8 companies of Korean stock market, and carried out the goodness of fit and independence tests. We found out the distributions of absolute returns are closed to compressed exponential distribution. The parameters are dominant that 1 < ${\beta}$ < 2, followed by ${\beta}=1$(exponential distribution) and ${\beta}=2$(normal distribution). Meanwhile, we assured that most of the absolute returns for major 8 companies have relevance to each other by chi-square independence test.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.10
• /
• pp.2310-2316
• /
• 1997

This paper presents a comparative performance analysis of the stochastic gradient adaptive algorithm based on the least mean absolute third (LMAT) error criterion with other widely-used competing adaptive algorithms. Under the assumption that the signals involved are zero-mean, wide-sense stationary and Gaussian, approximate expressions that characterize the steady-state mean-squared estimation error of the algorithm is dervied. The validity of our derivation is then confirement by computer simulations. The convergence speed is compared under the condition that the LMAT and other competing algorithms converge to the same value for the mean-squared estimation error in the stead-state, and superior convergence property of the LMAT algorithm is observed. In particular, it is shown that the LMAT algorithm converges faster than other algorithms even through the eignevalue spread ratio of the input signal and measurement noise power change.

• The Korean Journal of Applied Statistics
• /
• v.20 no.3
• /
• pp.531-540
• /
• 2007

The recursive feature elimination for support vector machine is known to be useful in selecting relevant genes. Since the criterion for choosing relevant genes is the absolute value of a coefficient, the recursive feature elimination may suffer from a scaling problem. We propose a modified version of the recursive feature elimination algorithm using bootstrap. In our method, the criterion for determining relevant genes is the absolute value of a coefficient divided by its standard error, which accounts for statistical variability of the coefficient. Through numerical examples, we illustrate that our method is effective in gene selection.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.4A
• /
• pp.693-700
• /
• 2006

To detect and evaluate the damage present in bridge, static identification method is known to be simple and effective, compared to dynamic method. In this study, the damage detection method in steel box girder bridge using static responses including displacement, slope and curvature is examined. The static displacement is calculated using finite element analysis and the slope and curvature are determined from the displacement using central difference method. The location of damage is detected using the absolute differences of these responses in intact and damaged bridge. Steel box girder bridge with corner crack is modeled using singular element in finite element method. The results show that these responses were significantly useful in detecting and predicting the location of damage present in bridge.

• Korean Chemical Engineering Research
• /
• v.62 no.3
• /
• pp.269-273
• /
• 2024

The normal stress of TiO₂ ER fluid under an electric field showed negative values due to the electrostatic attraction force in the normal direction between particles and the absolute value increased dramatically with electric field strengths. The normal yield stress exhibited E² dependence similar to the dynamic yield stress, indicating that normal stress can be utilized for evaluating the ER effect. Numerical simulation demonstrated good qualitative agreement with the experimental data and suggested that the decrease in the absolute value of normal stress with increasing shear rates was attributed to the rearrangement of particle configurations under shear.

• The Korean Journal of Financial Studies
• /
• v.12 no.1
• /
• pp.1-17
• /
• 2006

쇄신의 분산이 무한인 주가시계열이 장기의존성 과정에 의하여 생성되고 있는가 또는 생성되고 있지 않는가를 검정하고자 한다. 기존의 연구가 쇄신의 분산이 유한한 경우에 한정하여 장기의존성 주가 과정에 대한 장기기억성이 검토되어왔다. 이 논문에서는 쇄신의 분산이 유한한 경우와 무한한 경우에 다같이 적용되는 방법들을 한국종합주가지수의 일별수익률에 적용하여 장기기억 모수를 추정 검정한다. 추정방법으로서는 분수 가우스 잡음, 가우스 분수적분 자기회기 이동평균, 선형 분수안정잡음 등이 형성되는 상황에 절대값 방법, 분수 방법과 총량화 Whittle 방법을 사용한다. 한국종합주가지수의 일별대수수익률 시계열은 분산이 무한한 경우에도 장기의존성과정에 의하여 생성되고 있다. 극치가 존재해도 장기기억과정이 형성 되고 있다.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2000.04a
• /
• pp.141-144
• /
• 2000

스테레오 음향 반향 제거기는 채널신호간에 상호상관(cross-correlation) 때문에 적응필터 계수가 수신실의 반향 경로를 정확하게 추정하지 못한다. 본 논문에서는 상호상관을 줄이기 위해서 채널신호간 차의 절대값과 시변 감쇠상수를 이용하는 새로운 전처리 필터를 제안한다. 시뮬레이션을 통해서 제안하는 전처리 필터가 기존의 방법에 비해서 우수함을 보인다.