• Journal of the Korean Data and Information Science Society
• /
• 제9권1호
• /
• pp.63-70
• /
• 1998

The singular value decomposition is a tool which is used to find a linear structure of reduced dimension and to give interpretation of the lower dimensional structure about multivariate data. In this paper the singular value decomposition is reviewed from both algebraic and geometric point of view and, is illustrated the way which the tool is used in the multivariate techniques finding a simpler geometric structure for the data.

• Communications for Statistical Applications and Methods
• /
• 제28권5호
• /
• pp.553-562
• /
• 2021

Directional regression (DR; Li and Wang, 2007) is well-known as an exhaustive sufficient dimension reduction method, and performs well in complex regression models to have linear and nonlinear trends. However, the extension of DR is not well-done upto date, so we will extend DR to accommodate multivariate regression and large p-small n regression. We propose three versions of DR for multivariate regression and discuss how DR is applicable for the latter regression case. Numerical studies confirm that DR is robust to the number of clusters and the choice of hierarchical-clustering or pooled DR.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1994년도 제11회 음성통신 및 신호처리 워크샵 논문집 (SCAS 11권 1호)
• /
• pp.364-367
• /
• 1994

In this paper, we carried out some experiments on the Korean vowel recognition using the fractal dimension of the speech signals. We chose the Mincowski-Bouligand dimensioni as the fractal dimension, and computed it using the morphological covering method. For our experiments, we used both the fractal dimension and the LPC cepstrum which is conventionally known to be one of the best parameters for speech recognition, and examined the usefulness of the fractal dimension. From the vowel recognition experiments under various consonant contexts, we achieved the vowel recognition error rats of 5.6% and 3.2% for the case with only LPC cepstrum and that with both LPC cepstrum and the fractal dimension, respectively. The results indicate that the incorporation of the fractal dimension with LPC cepstrum gies more than 40% reduction in recognition errors, and indicates that the fractal dimension is a useful feature parameter for speech recognition.

• 응용통계연구
• /
• 제24권4호
• /
• pp.597-607
• /
• 2011

금융자산에의 투자에서 리스크 관리의 중요성이 부각되면서 리스크를 측정할 수 있는 도구로서 Value at Risk (VaR)가 널리 각광을 받고 있다. Value at Risk는 주어진 신뢰수준에서 목표기간 동안 발생 가능한 최대손실로 정의되는데 몇 가지 한계점이 있지만 비교적 간단하게 계산되고 이해될 수 있다는 장점이 있어 리스크 측정 및 관리의 기본적인 측도로 이용되고 있다. 그러나 포트폴리오에 포함되는 자산의 숫자가 많아지는 경우 VaR을 계산하는 데에 필수적인 변동성 추정이 매우 어려워지게 된다. 이때 차원축소의 방법을 생각할 수 있는데, 전통적인 인자분석은 시계열자료에 적합한 방법이 아니기 때문에 직접 적용할 수 없고 자료의 자기상관성을 제거하는 방법이 선행되어야 한다. 본 논문에서는 인자분석의 확장 형태인 시계열인자분석을 활용하여 시계열자료의 차원축소과정을 간결하게 하는 방법을 제시하고, 시계열인자분석으로 차원을 축소할 때 기존의 방법을 사용하는 것과 어떠한 차이가 있는지를 실제 금융자료를 이용한 VaR의 사후검증을 통해 분석하였다.

• 대한두개안면성형외과학회지
• /
• 제22권3호
• /
• pp.148-153
• /
• 2021

Background: The zygomaticomaxillary complex (ZMC) has a protruded, convex shape and plays a vital role in determining the contour by affecting the width of the middle face. This study aimed to evaluate the efficiency of ZMC fracture reduction and explore detailed directions for outcome improvement. Methods: We conducted a retrospective study of patients diagnosed with unilateral ZMC fracture who underwent ZMC reduction surgery at a single hospital between January 2015 and May 2020. The primary outcome variable was facial asymmetry using the difference in the bilateral malar eminence (ME) position measured by computed tomography scan. The 3-dimensional distance (I_A, asymmetry index) and the distance in each dimension, D_x (anteroposterior distance), D_y (mediolateral distance), and D_z (superoinferior distance) were compared. Results: A total of 101 patients with ZMC fractures and 54 non-fracture patients were enrolled in the study. The mean age of the study sample was 43.49 years (control sample, 43.35 years), and the male-to-female ratio was 66.3:33.7 (control sample, 64.8:35.2). There were 53 and 48 patients with right and left ZMC fractures, respectively. The I_A was not statistically different between the two groups. In terms of position in each dimension, only D_x was significantly different between the two groups. Conclusion: The results show that overall facial asymmetry was recovered after ZMC reduction, but in certain dimension significant difference in ME position has still remained. For further improvement, treatment should be performed to relieve malar depression in the anteroposterior dimension.

• 한국산업융합학회 논문집
• /
• 제5권1호
• /
• pp.33-38
• /
• 2002

This paper proposes an efficient method for improving the performance of polynomial adaline using the dimension reduction of independent variables. The adaptive principal component analysis is applied for reducing the dimension by extracting efficiently the features of the given independent variables. It can be solved the problems due to high dimensional input data in the polynomial adaline that the principal component analysis converts input data into set of statistically independent features. The proposed polynomial adaline has been applied to classify the patterns. The simulation results shows that the proposed polynomial adaline has better performances of the classification for test patterns, in comparison with those using the conventional polynomial adaline. Also, it is affected less by the scope of the smoothing factor.

• 한국CDE학회논문집
• /
• 제11권3호
• /
• pp.172-182
• /
• 2006

This paper presents a method to convert 3D CAD model to an appropriate analysis model using wrap-around, smooth-out and thinning operators that have been originally developed to realize the multi-resolution modeling. Wrap-around and smooth-out operators are used to simplify 3D model, and thinning operator is to reduce the dimension of a target object with simultaneously decomposing the simplified 3D model to 1D or 2D shapes. By using the simplification and dimension-reduction operations in an appropriate way, the user can generate an analysis model that matches specific applications. The advantage of this method is that the user can create optimized analysis models of various simplification levels by selecting appropriate number of detailed features and removing them.

• 농업과학연구
• /
• 제38권3호
• /
• pp.525-532
• /
• 2011

The aim of this study was to develop a portable electronic nose system for freshness measurement of meats, which could be an alterative of subjective measurements of human nose and time-consuming measurements of conventional gas chromatograph methods. The portable electronic system was o optimized by comparing the measurement sensitivity and hardware efficiency, such as power consumption and dimension reduction throughout two stages of the prototypes. The electronic nose systems were constructed using an array of four different metal oxide semiconductor sensors. Two different configurations of sensor array with dimension were designed and compared the performance respectively. The final prototype of the system showed much improved performance on saving power consumption and dimension reduction without decrease of measurement sensitivity of pork freshness. The results show the potential of constructing a portable electronic system for the measurement of meat quality with high sensitivity and energy efficiency.

• 한국콘텐츠학회:학술대회논문집
• /
• 한국콘텐츠학회 2006년도 추계 종합학술대회 논문집
• /
• pp.533-535
• /
• 2006

본 논문은 얼굴 인식에서 널리 사용되고 있는 PCA 기법에서 1, 2, 3차의 저차원의 특징 벡터를 배제하여 조명있는 얼굴의 인식률 변화를 실험하였다. 보편적으로 저차원 3개를 배제할 경우 조명에 강건한 얼굴 인식을 보인다고 하나, 저차원의 어느 부분이 조명에 크게 관여가되는지는 알려지지 않고 있다. 이에 본 연구에서는 1차, 2차, 3차 및 이를 조합하여 저차원의 조명에 대한 영향을 분석하였다.

• 전자공학회논문지
• /
• 제50권8호
• /
• pp.61-66
• /
• 2013

본 논문에서는 채널부호화 다중 안테나 시스템에서 공간다중화 전송된 신호들을 효과적으로 복조하기 위한 저연산 연판정 복조 다중 안테나 반복검출 기법을 제시한다. 반복 검출기법의 경우 우수한 성능에도 불구하고 연산량의 복잡성으로 수신단에 높은 복잡도를 요청하게 된다. 이러한 복잡도 감소를 위해 차원감소 소프트 검출 기법 (DRSD)과 모든 순서 순차적 간섭 제거(AOSIC) 기법을 사용한다. 이 기법의 경우 기존 기법들에 비해 반복검출 기법의 연산량의 복잡성을 줄일 수 있으며 향상된 성능을 얻을 수 있다.